Search results for: segmentation of aggregating superpixels

Authors: Hongbo Zhang, Xinlu Tang, Jiangwei Li, Chi Yan

Abstract:

Home zone parking pilot (HPP) is a fast-growing segment in low-speed autonomous driving applications. It requires the car automatically cruise around a parking lot and park itself in a range of up to 100 meters inside a recurrent home/office parking lot, which requires precise parking lot mapping and localization solution. Although Lidar is ideal for SLAM, the car OEMs favor a low-cost fish-eye camera based visual SLAM approach. Recent approaches have employed segmentation models to extract semantic features and improve mapping accuracy, but these AI models are memory unfriendly and computationally expensive, making deploying on embedded ADAS systems difficult. To address this issue, we proposed a new method that utilizes object detection models to extract robust and accurate parking lot features. The proposed method could reduce computational costs while maintaining high accuracy. Once combined with vehicles’ wheel-pulse information, the system could construct maps and locate the vehicle in real-time. This article will discuss in detail (1) the fish-eye based Around View Monitoring (AVM) with transparent chassis images as the inputs, (2) an Object Detection (OD) based feature point extraction algorithm to generate point cloud, (3) a low computational parking lot mapping algorithm and (4) the real-time localization algorithm. At last, we will demonstrate the experiment results with an embedded ADAS system installed on a real car in the underground parking lot.

Keywords: ADAS, home zone parking pilot, object detection, visual SLAM

Procedia PDF Downloads 67

Authors: Eid Alotaibi

Abstract:

The purpose of this study is to investigate the impact of tourists’ culture on perception and evaluation of hotel service experience and behavioral intentions. Drawing on Hofested’s cultural dimensions, this study seeks to further contribute towards understanding the effect of culture on perception and evaluation of hotels’ services, and whether there are differences between Saudi and European tourists’ perceptions of hotel services evaluation. A descriptive cross-sectional design was used in this study. Data were collected from tourists staying in five-star hotels in Saudi Arabia using the self-completion technique. The findings show that evaluations of hotel services differ from one culture to another. T-test results reveal that Saudis were more tolerant and reported significantly higher levels of satisfaction, were more likely to return and recommend the hotel, and perceived the price for the hotel stay as being good value for money as compared to their European counterparts. The sample was relatively small and specific to only five-star hotel evaluations. As a result, findings cannot be generalized to the wider tourist population. The results of this research have important implications for management within the Saudi hospitality industry. The study contributes to the tourist cultural theory by emphasizing the relative importance of cultural dimensions in-service evaluation. The author argues that no studies could be identified that compare Saudis and Europeans in their evaluations of their experiences staying at hotels. Therefore, the current study would enhance understanding of the effects of cultural factors on service evaluations and provide valuable input for international market segmentation and resource allocation in the Saudi hotel industry.

Keywords: culture, tourist, service experience, hotel industry, Hofested’s cultural dimensions

Procedia PDF Downloads 119

Authors: Juan de Ona, Esperanza Estevez, Rocío de Ona

Abstract:

Urban traffic levels can be reduced by drawing travelers away from private vehicles over to using public transport. This modal change can be achieved by either introducing restrictions on private vehicles or by introducing measures which increase people’s satisfaction with public transport. For public transport users, quality of service affects customer satisfaction, which, in turn, influences the behavioral intentions towards the service. This paper intends to identify the main attributes which influence the perception private vehicle users have about the public transport services provided in five European cities: Berlin, Lisbon, London, Madrid and Rome. Ordinal logit models have been applied to an online panel survey with a sample size of 2,500 regular private vehicle users (approximately 500 inhabitants per city). To achieve a comprehensive analysis and to deal with heterogeneity in perceptions, 15 models have been developed for the entire sample and 14 user segments. The results show differences between the cities and among the segments. Madrid was taken as reference city and results indicate that the inhabitants are satisfied with public transport in Madrid and that the most important public transport service attributes for private vehicle users are frequency, speed and intermodality. Frequency is an important attribute for all the segments, while speed and intermodality are important for most of the segments. An analysis by segments has identified attributes which, although not important in most cases, are relevant for specific segments. This study also points out important differences between the five cities. Findings from this study can be used to develop policies and recommendations for persuading.

Keywords: service quality, satisfaction, public transportation, private vehicle users, car users, segmentation, ordered logit

Procedia PDF Downloads 117

Authors: Duvert A. Puentes T., Javier A. Maldonado E., Ivan Quintero., Diego F. Villegas

Abstract:

Mechanical Computation is a great tool to study the performance of complex models. An example of it is the study of the human body structure. This paper took advantage of different types of software to make a 3D model of the glenohumeral joint and apply a finite element analysis. The main objective was to study the change in the biomechanical properties of the joint when it presents an injury. Specifically, a BANKART lesion, which consists in the detachment of the anteroinferior labrum from the glenoid. Stress and strain distribution of the soft tissues were the focus of this study. First, a 3D model was made of a joint without any pathology, as a control sample, using segmentation software for the bones with the support of medical imagery and a cadaveric model to represent the soft tissue. The joint was built to simulate a compression and external rotation test using CAD to prepare the model in the adequate position. When the healthy model was finished, it was submitted to a finite element analysis and the results were validated with experimental model data. With the validated model, it was sensitized to obtain the best mesh measurement. Finally, the geometry of the 3D model was changed to imitate a BANKART lesion. Then, the contact zone of the glenoid with the labrum was slightly separated simulating a tissue detachment. With this new geometry, the finite element analysis was applied again, and the results were compared with the control sample created initially. With the data gathered, this study can be used to improve understanding of the labrum tears. Nevertheless, it is important to remember that the computational analysis are approximations and the initial data was taken from an in vitro assay.

Keywords: biomechanics, computational model, finite elements, glenohumeral joint, bankart lesion, labrum

Procedia PDF Downloads 161

Authors: Bryce R. Jolley, Michael Uchic

Abstract:

This study presents initial results toward the correlative characterization of inherent defects of Ti-6Al-4V additive manufacture (AM). X-Ray Computed Tomography (CT) defect data are compared and correlated with microscopic photographs obtained via automated serial sectioning. The metal AM specimen was manufactured out of Ti-6Al-4V virgin powder to specified dimensions. A post-contour was applied during the fabrication process with a speed of 1050 mm/s, power of 260 W, and a width of 140 µm. The specimen was stress relief heat-treated at 16°F for 3 hours. Microfocus CT imaging was accomplished on the specimen within a predetermined region of the build. Microfocus CT imaging was conducted with parameters optimized for Ti-6Al-4V additive manufacture. After CT imaging, a modified RoboMet. 3D version 2 was employed for serial sectioning and optical microscopy characterization of the same predetermined region. Automated montage capture with sub-micron resolution, bright-field reflection, 12-bit monochrome optical images were performed in an automated fashion. These optical images were post-processed to produce 2D and 3D data sets. This processing included thresholding and segmentation to improve visualization of defect features. The defects observed from optical imaging were compared and correlated with the defects observed from CT imaging over the same predetermined region of the specimen. Quantitative results of area fraction and equivalent pore diameters obtained via each method are presented for this correlation. It is shown that Microfocus CT imaging does not capture all inherent defects within this Ti-6Al-4V AM sample. Best practices for this correlative effort are also presented as well as the future direction of research resultant from this current study.

Keywords: additive manufacture, automated serial sectioning, computed tomography, nondestructive evaluation

Procedia PDF Downloads 141

Authors: Nasibeh Mohammadpour, Ali Rajabzadeh, Adel Azar, Hamid Zargham Borujeni,

Abstract:

Generally, various industries development depends on their stakeholders and beneficiaries supports. One of the most important stakeholders in tourism industry ( which has become one of the most important lucrative and employment-generating activities at the international level these days) are host communities in tourist destination which are affected and effect on this industry development. Recognizing host community and its segmentations can be important to get their support for future decisions and policy making. In order to identify these segments, in this study, clustering of the residents has been done by using some tools that are designed to encounter human complexities and have ability to model and generalize complex systems without any needs for the initial clusters’ seeds like classic methods. Neural networks can help to meet these expectations. The research have been planned to design neural networks-based mathematical model for clustering the host community effectively according to multi criteria, and identifies differences among segments. In order to achieve this goal, the residents’ segmentation has been done by demographic characteristics, their attitude towards the tourism development, the level of satisfaction and the type of their support in this field. The applied method is self-organized neural networks and the results have compared with K-means. As the results show, the use of Self- Organized Map (SOM) method provides much better results by considering the Cophenetic correlation and between clusters variance coefficients. Based on these criteria, the host community is divided into five sections with unique and distinctive features, which are in the best condition (in comparison other modes) according to Cophenetic correlation coefficient of 0.8769 and between clusters variance of 0.1412.

Keywords: Artificial Nural Network, Clustering , Resident, SOM, Tourism

Procedia PDF Downloads 183

Authors: Levent Basayigit, Sinan Demir, Burhan Kara, Yusuf Ucar

Abstract:

Cannabis is mostly used for drug production. In some countries, an excessive amount of illegal cannabis is cultivated and sold. Most of the illegal cannabis cultivation occurs on the lands far from settlements. In farmlands, it is cultivated with other crops. In this method, cannabis is surrounded by tall plants like corn and sunflower. It is also cultivated with tall crops as the mixed culture. The common method of the determination of the illegal cultivation areas is to investigate the information obtained from people. This method is not sufficient for the determination of illegal cultivation in remote areas. For this reason, more effective methods are needed for the determination of illegal cultivation. Remote Sensing is one of the most important technologies to monitor the plant growth on the land. The aim of this study is to monitor cannabis cultivation area using satellite imagery. The main purpose of this study was to develop an applicable method for monitoring the cannabis cultivation. For this purpose, cannabis was grown as single or surrounded by the corn and sunflower in plots. The morphological characteristics of cannabis were recorded two times per month during the vegetation period. The spectral signature library was created with the spectroradiometer. The parcels were monitored with high-resolution satellite imagery. With the processing of satellite imagery, the cultivation areas of cannabis were classified. To separate the Cannabis plots from the other plants, the multiresolution segmentation algorithm was found to be the most successful for classification. WorldView Improved Vegetative Index (WV-VI) classification was the most accurate method for monitoring the plant density. As a result, an object-based classification method and vegetation indices were sufficient for monitoring the cannabis cultivation in multi-temporal Earthwiev images.

Keywords: Cannabis, drug, remote sensing, object-based classification

Procedia PDF Downloads 272

Authors: Natascha C. D'Amico, Enzo Grossi, Giovanni Valbusa, Ala Malasevschi, Gianpiero Cardone, Sergio Papa

Abstract:

Purpose: To characterize, through a radiomic approach, the nature of areas classified PI-RADS (Prostate Imaging Reporting and Data System) 3/5, recognized in multiparametric prostate magnetic resonance with T2-weighted (T2w), diffusion and perfusion sequences with paramagnetic contrast. Methods and Materials: 24 cases undergoing multiparametric prostate MR and biopsy were admitted to this pilot study. Clinical outcome of the PI-RADS 3/5 was found through biopsy, finding 8 malignant tumours. The analysed images were acquired with a Philips achieva 1.5T machine with a CE- T2-weighted sequence in the axial plane. Semi-automatic tumour segmentation was carried out on MR images using 3DSlicer image analysis software. 45 shape-based, intensity-based and texture-based features were extracted and represented the input for preprocessing. An evolutionary algorithm (a TWIST system based on KNN algorithm) was used to subdivide the dataset into training and testing set and select features yielding the maximal amount of information. After this pre-processing 20 input variables were selected and different machine learning systems were used to develop a predictive model based on a training testing crossover procedure. Results: The best machine learning system (three-layers feed-forward neural network) obtained a global accuracy of 90% ( 80 % sensitivity and 100% specificity ) with a ROC of 0.82. Conclusion: Machine learning systems coupled with radiomics show a promising potential in distinguishing benign from malign tumours in PI-RADS 3/5 areas.

Keywords: machine learning, MR prostate, PI-Rads 3, radiomics

Procedia PDF Downloads 188

Authors: Nur Nadhirah Rusyda Rosnan, Nursuhaili Najwa Masrol, Nurul Fatiha MD Nor, Mohammad Zafrullah Mohammad Salim, Sim Choon Cheak

Abstract:

Accurate inventories of oil palm planted areas are crucial for plantation management as this would impact the overall economy and production of oil. One of the technological advancements in the oil palm industry is semi-automated palm counting, which is replacing conventional manual palm counting via digitizing aerial imagery. Most of the semi-automated palm counting method that has been developed was limited to mature palms due to their ideal canopy size represented by satellite image. Therefore, immature palms were often left out since the size of the canopy is barely visible from satellite images. In this paper, an approach using a morphological filter and high-resolution satellite image is proposed to detect immature palm trees. This approach makes it possible to count the number of immature oil palm trees. The method begins with an erosion filter with an appropriate window size of 3m onto the high-resolution satellite image. The eroded image was further segmented using watershed segmentation to delineate immature palm tree regions. Then, local minimum detection was used because it is hypothesized that immature oil palm trees are located at the local minimum within an oil palm field setting in a grayscale image. The detection points generated from the local minimum are displaced to the center of the immature oil palm region and thinned. Only one detection point is left that represents a tree. The performance of the proposed method was evaluated on three subsets with slopes ranging from 0 to 20° and different planting designs, i.e., straight and terrace. The proposed method was able to achieve up to more than 90% accuracy when compared with the ground truth, with an overall F-measure score of up to 0.91.

Keywords: immature palm count, oil palm, precision agriculture, remote sensing

Procedia PDF Downloads 76

Authors: Chandu Rathnayake, Isuri Anuradha

Abstract:

Early detection and accurate diagnosis of lung diseases play a crucial role in improving patient prognosis. However, conventional diagnostic methods heavily rely on subjective symptom assessments and medical imaging, often causing delays in diagnosis and treatment. To overcome this challenge, we propose a novel lung disease prediction system that integrates patient symptoms and X-ray images to provide a comprehensive and reliable diagnosis.In this project, develop a mobile application specifically designed for detecting lung diseases. Our application leverages both patient symptoms and X-ray images to facilitate diagnosis. By combining these two sources of information, our application delivers a more accurate and comprehensive assessment of the patient's condition, minimizing the risk of misdiagnosis. Our primary aim is to create a user-friendly and accessible tool, particularly important given the current circumstances where many patients face limitations in visiting healthcare facilities. To achieve this, we employ several state-of-the-art algorithms. Firstly, the Decision Tree algorithm is utilized for efficient symptom-based classification. It analyzes patient symptoms and creates a tree-like model to predict the presence of specific lung diseases. Secondly, we employ the Random Forest algorithm, which enhances predictive power by aggregating multiple decision trees. This ensemble technique improves the accuracy and robustness of the diagnosis. Furthermore, we incorporate a deep learning model using Convolutional Neural Network (CNN) with the RestNet50 pre-trained model. CNNs are well-suited for image analysis and feature extraction. By training CNN on a large dataset of X-ray images, it learns to identify patterns and features indicative of lung diseases. The RestNet50 architecture, known for its excellent performance in image recognition tasks, enhances the efficiency and accuracy of our deep learning model. By combining the outputs of the decision tree-based algorithms and the deep learning model, our mobile application generates a comprehensive lung disease prediction. The application provides users with an intuitive interface to input their symptoms and upload X-ray images for analysis. The prediction generated by the system offers valuable insights into the likelihood of various lung diseases, enabling individuals to take appropriate actions and seek timely medical attention. Our proposed mobile application has significant potential to address the rising prevalence of lung diseases, particularly among young individuals with smoking addictions. By providing a quick and user-friendly approach to assessing lung health, our application empowers individuals to monitor their well-being conveniently. This solution also offers immense value in the context of limited access to healthcare facilities, enabling timely detection and intervention. In conclusion, our research presents a comprehensive lung disease prediction system that combines patient symptoms and X-ray images using advanced algorithms. By developing a mobile application, we provide an accessible tool for individuals to assess their lung health conveniently. This solution has the potential to make a significant impact on the early detection and management of lung diseases, benefiting both patients and healthcare providers.

Keywords: CNN, random forest, decision tree, machine learning, deep learning

Procedia PDF Downloads 73

Authors: Spyridon Komaitis, Christos Koutsarnakis, Evangelos Drosos, Aristotelis Kalyvas

Abstract:

This study opts to investigate the intrinsic architecture, morphology, and spatial relationship of the subcortical pathways implicated in the connectivity of the motor/premotor cortex and SMA/pre-SMA complex. Twenty normal, adult, formalin-fixed cerebral hemispheres were explored through the fiber micro-dissection technique. Lateral to medial and medial to lateral dissections focused on the area of interest were performed in a tandem manner and under the surgical microscope. We traced the subcortical architecture, spatial relationships, and axonal connectivity of four major pathways: a) the dorsal component of the SLF (SLF-I) was found to reside in the medial aspect of the hemisphere and seen to connect the precuneus with the SMA and pre-SMA complex, b) the frontal longitudinal system (FLS) was consistently encountered as the natural anterior continuation of the SLF-II and SLF-III and connected the premotor and prefrontal cortices c) the fronto-caudate tract (FCT), a fan-shaped tract, was documented to participate in connectivity of the prefrontal and premotor cortices to the head and body of the caudate nucleus and d) the cortico-tegmental tract(CTT) was invariably recorded to subserve the connectivity of the tegmental area with the fronto-parietal cortex. No hemispheric asymmetries were recorded for any of the implicated pathways. Sub-segmentation systems were also proposed for each of the aforementioned tracts. The structural connectivity and functional specialization of motor and premotor areas in the human brain remain vague to this day as most of the available evidence derives either from animal or tractographic studies. By using the fiber-microdissection technique as our main method of investigation, we provide sound structural evidence on the delicate anatomy of the related white matter pathways.

Keywords: neuroanatomy, premotor, motor, connectivity

Procedia PDF Downloads 128

Authors: Vaishali Joshi, Pallav Joshi

Abstract:

The main objective of this study is to examine the fashion consumption behavior of females with respect to their age group. Differences were studied in the pre-purchase, purchase and post-purchase behavior of females belonging to three age cohorts such as Generation X, Generation Y, and Generation Z. Quantitative approach was used to conduct this research. Data was collected through structured questionnaire. The questionnaire consisted of three sections. Section one included a question of the source of information of purchasing fashion apparels which measure the pre-purchase behavior. Section two measures purchase behavior which included two questions: i. motivations for purchasing fashion apparel and ii. important attributes considered for purchasing fashion apparel. The last section included a question regarding disposal of fashion apparel which measures the post-purchase behavior. Hundred females were selected as the respondents for this study through convenience sampling in the fashion streets. They were categorized into three age groups and then the results were analyzed. Four hypotheses were developed after reviewing the existing literature. Regression analysis was conducted for testing the hypothesis. Hypothesis one was accepted which stated that ‘social influence’ as a source of information for purchasing fashion apparels decreases with age. Hypothesis two was accepted which suggested that motivation of ‘Attention seeking’ for purchasing fashion apparel decreases with age. Hypothesis three and four also accepted which suggested that the importance of ‘Quality’ and ‘Price’ increases with age but hypothesis five was rejected which suggested that the importance of ‘Fit’ increases with age and last but not the least hypothesis six was accepted which suggested that the ‘duration’ of using fashion apparel increases with age. Limitation of the study deals with the sample of only female respondents. Implication can be made from this research in the field of Fashion apparel industry with respect to consumer segmentation and better marketing approaches can be implemented by the marketers form this study. Further research can be concluded by including male respondents also.

Keywords: fashion, consumption behavior, age cohorts, motivation

Procedia PDF Downloads 267

Authors: Shu-Yuan Lin, Tung-Chin Chou

Abstract:

As products become increasingly similar in competitive markets, achieving product segmentation and differentiation through packaging design has become the primary task when designing retail product packaging. When the main focus of brand marketing is no longer the product itself, emotional marketing, such as the use of humorous packaging designs, may be employed to successfully promote the brand. Such efforts will capture the hearts of consumers, generate discussions, and allow the brand to leave a deep impression in consumers. In this study, snack packaging was used to develop a relationship model that illustrated the effect of humorous packaging designs on brand awareness and brand attitude. The study was divided into three stages: In the first stage, in-depth interviews and focus group interviews were conducted with experts to construct 24 indicators for assessing humorous packaging designs. In the second stage, survey questionnaires were distributed to a young consumer group; the results showed that the group had a high and low product involvement with chocolate and dried shredded squid, respectively. Humorous packaging designs were subsequently created for two snack types to produce a study sample of 12 different packaging. In the third stage, packaging designs were evaluated by obtaining scores for the consumers’ brand awareness, brand attitude, and perceived effects of the packaging designs. Finally, a relationship model was developed to show the effect of humorous packaging designs on brand awareness and brand attitude, confirming that two perceived effects of humorous packaging designs (i.e., ‘pleasant and emotionally healing’ and ‘connected to people’s daily life’) exhibited a significant and positive effect on ‘perceived brand value,’ where the effect of ‘pleasant and emotionally healing’ was the most significant. In addition, ‘pleasant and emotionally healing’ exerted a significant and positive effect on ‘brand purchase intention.’ Furthermore, packaging designs with humorous elements helped foster brand awareness.

Keywords: brand awareness, brand attitude, humorous design, packaging design

Procedia PDF Downloads 226

Authors: Marthinus Conradie

Abstract:

The conversational dynamics of democratically orientated deliberation continue to stimulate critical scholarship for its potential to bolster robust engagement between different sections of pluralist societies. Several axes of deliberation that have attracted academic attention include face-to-face vs. online interaction, and citizen-to-citizen communication vs. engagement between citizens and political elites. In all these areas, numerous researchers have explored deliberative procedures aimed at achieving instrumental goals such a securing consensus on policy issues, against procedures that prioritise expressive outcomes such as broadening the range of argumentative repertoires that discursively construct and mediate specific political issues. The study that informs this paper, works in the latter stream. Drawing its data from the reader-comments section of a South African broadsheet newspaper, the study investigates online, citizen-to-citizen deliberation by analysing the discursive practices through which competing understandings of social problems are articulated and contested. To advance this agenda, the paper deals specifically with user-generated comments posted in response to news stories on questions of race and racism in South Africa. The analysis works to discern and interpret the various sets of discourse practices that shape how citizens deliberate contentious political issues, especially racism. Since the website in question is designed to encourage the critical comparison of divergent interpretations of news events, without feeding directly into national policymaking, the study adopts an analytic framework that traces how citizens articulate arguments, rather than the instrumental effects that citizen deliberations might exert on policy. The paper starts from the argument that such expressive interactions are particularly crucial to current trends in South African politics, given that the precise nature of race and racism remain contested and uncertain. Centred on a sample of 2358 conversational moves in 814 posts to 18 news stories emanating from issues of race and racism, the analysis proceeds in a two-step fashion. The first stage conducts a qualitative content analysis that offers insights into the levels of reciprocity among commenters (do readers engage with each other or simply post isolated opinions?), as well as the structures of argumentation (do readers support opinions by citing evidence?). The second stage involves a more fine-grained discourse analysis, based on a theorisation of argumentation that delineates it into three components: opinions/conclusions, evidence/data to support opinions/conclusions and warrants that explicate precisely how evidence/data buttress opinions/conclusions. By tracing the manifestation and frequency of specific argumentative practices, this study contributes to the archive of research currently aggregating around the practices that characterise South Africans’ engagement with provocative political questions, especially racism and racial inequity. Additionally, the study also contributes to recent scholarship on the affordances of Web 2.0 software by eschewing a simplistic bifurcation between cyber-optimist vs. pessimism, in favour of a more nuanced and context-specific analysis of the patterns that structure online deliberation.

Keywords: online deliberation, discourse analysis, qualitative content analysis, racism

Procedia PDF Downloads 177

Authors: Kakhaber Djakeli

Abstract:

For market leadership, iconic brands already start to establish their unique digital avatars into Metaverse and offer Non Fungible Tokens to their fans. Metaverse can be defined as an evolutionary step of Internet development. So if companies and brands use the internet, logically, they can find new solutions for them and their customers in Metaverse. Marketing and Management today must learn how to combine physical world activities with those either entitled as digital, virtual, and immersive. A “Phygital” Solution uniting physical and digital competitive activities of the company covering the questions about how to use virtual worlds for Brand Development and Non Fungible Tokens for more attractiveness soon will be most relevant question for Branding. Thinking comprehensively, we can entitle this type of branding as an Immersive one. As we see, the Immersive Brands give customers more mesmerizing feelings than traditional ones. Accordingly, the Branding can be divided by the company in its own understanding into two models: traditional and immersive. Immersive Branding being more directed to Sensorial challenges of Humans will be big job for International Universities in near future because they target the Generation - Z. To try to help those International Universities opening the door to the mesmerizing, immersive branding, the Marketing Research have been undertaken. The main goal of the study was to establish the model for Immersive Branding at International Universities and answer on many questions what logically arises in university life. The type of Delphi Surveys entitled as an Expert Studies was undertaken for one great mission, to help International Universities to open the opportunities to Phygital activities with reliable knowledge with Model of Immersive Branding. The Questionnaire sent to Experts of Education were covering professional type of questions from education to segmentation of customers, branding, attitude to students, and knowledge to Immersive Marketing. The research results being very interesting and encouraging enough to make author to establish the New Model of Immersive Experiential Branding for International Universities.

Keywords: branding, immersive marketing, students, university

Procedia PDF Downloads 81

Authors: Jou-An Chung

Abstract:

The purpose of this study is to investigate the verb bias of the Mandarin verbs in children’s reading materials and provide the criteria for categorization. Verb bias varies cross-linguistically. As Mandarin and English are typological different, this study hopes to shed light on Mandarin verb bias with the use of corpus and provide thorough and detailed criteria for analysis. Moreover, this study focuses on children’s reading materials since it is a significant issue in understanding children’s sentence processing. Therefore, investigating verb bias of Mandarin verbs in children’s reading materials is also an important issue and can provide further insights into children’s sentence processing. The small corpus is built up for this study. The corpus consists of the collection of school textbooks and Mandarin Daily News for children. The files are then segmented and POS tagged by JiebaR (Chinese segmentation with R). For the ease of analysis, the one-word character verbs and intransitive verbs are excluded beforehand. The total of 20 high frequency verbs are hand-coded and are further categorized into one of the three types, namely DO type, SC type and other category. If the frequency of taking Other Type exceeds the threshold of 25%, the verb is excluded from the study. The results show that 10 verbs are direct object bias verbs, and six verbs are sentential complement bias verbs. The paired T-test was done to assure the statistical significance (p = 0.0001062 for DO bias verb, p=0.001149 for SC bias verb). The result has shown that in children’s reading materials, the DO biased verbs are used more than the SC bias verbs since the simplest structure of sentences is easier for children’s sentence comprehension or processing. In sum, this study not only discussed verb bias in child's reading materials but also provided basic coding criteria for verb bias analysis in Mandarin and underscored the role of context. Sentences are easier for children’s sentence comprehension or processing. In sum, this study not only discussed verb bias in child corpus, but also provided basic coding criteria for verb bias analysis in Mandarin and underscored the role of context.

Keywords: corpus linguistics, verb bias, child language, psycholinguistics

Procedia PDF Downloads 291

Authors: Richard O’Riordan, Saritha Unnikrishnan

Abstract:

Autonomous Vehicles (AVs) are incorporating an increasing number of ADAS features, including automated lane-keeping systems. In recent years, many research papers into lane detection algorithms have been published, varying from computer vision techniques to deep learning methods. The transition from lower levels of autonomy defined in the SAE framework and the progression to higher autonomy levels requires increasingly complex models and algorithms that must be highly reliable in their operation and functionality capacities. Furthermore, these algorithms have no room for error when operating at high levels of autonomy. Although the current research details existing computer vision and deep learning algorithms and their methodologies and individual results, the research also details challenges faced by the algorithms and the resources needed to operate, along with shortcomings experienced during their detection of lanes in certain weather and lighting conditions. This paper will explore these shortcomings and attempt to implement a lane detection algorithm that could be used to achieve improvements in AV lane detection systems. This paper uses a pre-trained LaneNet model to detect lane or non-lane pixels using binary segmentation as the base detection method using an existing dataset BDD100k followed by a custom dataset generated locally. The selected roads will be modern well-laid roads with up-to-date infrastructure and lane markings, while the second road network will be an older road with infrastructure and lane markings reflecting the road network's age. The performance of the proposed method will be evaluated on the custom dataset to compare its performance to the BDD100k dataset. In summary, this paper will use Transfer Learning to provide a fast and robust lane detection algorithm that can handle various road conditions and provide accurate lane detection.

Keywords: ADAS, autonomous vehicles, deep learning, LaneNet, lane detection

Procedia PDF Downloads 104

Authors: Nawt Almutairi, Ramzi El-Haddadeh

Abstract:

With the extensive increase in the number of online shops, loyalty becomes the most purpose for e-retailers by which they can maintain their exit customers and regular income instead of spending large deal of money to target new segmentation. To obtain customers’ loyalty e-marketers should firstly satisfy customers by providing a high quality of services that could fulfil their demand. They have to satisfy them to trust the web-site then increase their intention to re-visit it. This study intends to investigate to what extend the elements of e-service quality presented in the literature affect customers’ satisfaction and how these influences contribute to customers’ trust and loyalty. Three dimensions of service quality are estimated. The first element is web-site interactivity, which is perceived the quality of interactive support and the accessible communications-tool. The second aspect is security/privacy, which is perceived the quality of controlling security and privacy while transaction over the web-site. The third element is web-design that perceived a pleasant user interface with visual appealing. These elements present positive effects on shoppers’ satisfaction. Thus, To examine the proposed constructs of this research, some measurements scale-items adapted from similar prior studies. Survey data collected online from Saudi customers (n=106) were utilized to test the research hypotheses. After that, the hypotheses were analyzed by using a variety of regression tools. The analytical results of this study propose that perceived quality of interactivity and security/privacy affects customers’ satisfaction. As well as trust seems to be a substantial construct that highly affects loyalty in online shopping. This study provides a developed model to obtain a simple understanding of the series of customers’ loyalty in online shopping. One construct presenting in the research model is web-design appears to be not important antecedent of satisfaction (the path to loyalty) in online shopping.

Keywords: e-service, satisfaction, trust, loyalty

Procedia PDF Downloads 259

Authors: N. Casado-Sanz, B. Guirao

Abstract:

The population aged 65 and over is projected to double in the coming decades. Due to this increase, driver population is expected to grow and in the near future, all countries will be faced with population aging of varying intensity and in unique time frames. This is the greatest challenge facing industrialized nations and due to this fact, the study of the relationships of dependency between population aging and road safety is becoming increasingly relevant. Although the deterioration of driving skills in the elderly has been analyzed in depth, to our knowledge few research studies have focused on the road infrastructure and the mobility of this particular group of users. In Spain, crosstown roads have one of the highest fatality rates. These rural routes have a higher percentage of elderly people who are more dependent on driving due to the absence or limitations of urban public transportation. Analysing road safety in these routes is very complex because of the variety of the features, the dispersion of the data and the complete lack of related literature. The objective of this paper is to identify key factors that cause traffic accidents. The individuals under study were the accidents with killed or seriously injured in Spanish crosstown roads during the period 2006-2015. Latent cluster analysis was applied as a preliminary tool for segmentation of accidents, considering population aging as the main input among other socioeconomic indicators. Subsequently, a linear regression analysis was carried out to estimate the degree of dependence between the accident rate and the variables that define each group. The results show that segmenting the data is very interesting and provides further information. Additionally, the results revealed the clear influence of the aging variable in the clusters obtained. Other variables related to infrastructure and mobility levels, such as the crosstown roads layout and the traffic intensity aimed to be one of the key factors in the causality of road accidents.

Keywords: cluster analysis, population ageing, rural roads, road safety

Procedia PDF Downloads 111

Authors: Morten Brøgger, Kim Wittchen

Abstract:

Focus on reducing energy consumption in existing buildings at large scale, e.g. in cities or countries, has been increasing in recent years. In order to reduce energy consumption in existing buildings, political incentive schemes are put in place and large scale investments are made by utility companies. Prioritising these investments requires a comprehensive overview of the energy consumption in the existing building stock, as well as potential energy-savings. However, a building stock comprises thousands of buildings with different characteristics making it difficult to model energy consumption accurately. Moreover, the complexity of the building stock makes it difficult to convey model results to policymakers and other stakeholders. In order to manage the complexity of the building stock, building archetypes are often employed in building stock energy models (BSEMs). Building archetypes are formed by segmenting the building stock according to specific characteristics. Segmenting the building stock according to building type and building age is common, among other things because this information is often easily available. This segmentation makes it easy to convey results to non-experts. However, using a single archetypical building to represent all buildings in a segment of the building stock is associated with loss of detail. Thermal characteristics are aggregated while other characteristics, which could affect the energy efficiency of a building, are disregarded. Thus, using a simplified representation of the building stock could come at the expense of the accuracy of the model. The present study evaluates the accuracy of a conventional archetype-based BSEM that segments the building stock according to building type- and age. The accuracy is evaluated in terms of the archetypes’ ability to accurately emulate the average energy demands of the corresponding buildings they were meant to represent. This is done for the buildings’ energy demands as a whole as well as for relevant sub-demands. Both are evaluated in relation to the type- and the age of the building. This should provide researchers, who use archetypes in BSEMs, with an indication of the expected accuracy of the conventional archetype model, as well as the accuracy lost in specific parts of the calculation, due to use of the archetype method.

Keywords: building stock energy modelling, energy-savings, archetype

Procedia PDF Downloads 154

Authors: Hritwik Ghosh, Irfan Sadiq Rahat, Sachi Nandan Mohanty, J. V. R. Ravindra

Abstract:

In the rapidly evolving landscape of medical diagnostics, the early detection and accurate classification of skin cancer remain paramount for effective treatment outcomes. This research delves into the transformative potential of Artificial Intelligence (AI), specifically Deep Learning (DL), as a tool for discerning and categorizing various skin conditions. Utilizing a diverse dataset of 3,000 images representing nine distinct skin conditions, we confront the inherent challenge of class imbalance. This imbalance, where conditions like melanomas are over-represented, is addressed by incorporating class weights during the model training phase, ensuring an equitable representation of all conditions in the learning process. Our pioneering approach introduces a hybrid model, amalgamating the strengths of two renowned Convolutional Neural Networks (CNNs), VGG16 and ResNet50. These networks, pre-trained on the ImageNet dataset, are adept at extracting intricate features from images. By synergizing these models, our research aims to capture a holistic set of features, thereby bolstering classification performance. Preliminary findings underscore the hybrid model's superiority over individual models, showcasing its prowess in feature extraction and classification. Moreover, the research emphasizes the significance of rigorous data pre-processing, including image resizing, color normalization, and segmentation, in ensuring data quality and model reliability. In essence, this study illuminates the promising role of AI and DL in revolutionizing skin cancer diagnostics, offering insights into its potential applications in broader medical domains.

Keywords: artificial intelligence, machine learning, deep learning, skin cancer, dermatology, convolutional neural networks, image classification, computer vision, healthcare technology, cancer detection, medical imaging

Procedia PDF Downloads 87

Authors: Justin Woulfe

Abstract:

Siloed logistics and supply chain management systems throughout the Department of Defense (DOD) has led to disparate approaches to modeling and simulation (M&S), a lack of understanding of how one system impacts the whole, and issues with “optimal” solutions that are good for one organization but have dramatic negative impacts on another. Many different systems have evolved to try to understand and account for uncertainty and try to reduce the consequences of the unknown. As the DoD undertakes expansive digital transformation initiatives, there is an opportunity to fuse and leverage traditionally disparate data into a centrally hosted source of truth. With a streamlined process incorporating machine learning (ML) and artificial intelligence (AI), advanced M&S will enable informed decisions guiding program success via optimized operational readiness and improved mission success. One of the current challenges is to leverage the terabytes of data generated by monitored systems to provide actionable information for all levels of users. The implementation of a cloud-based application analyzing data transactions, learning and predicting future states from current and past states in real-time, and communicating those anticipated states is an appropriate solution for the purposes of reduced latency and improved confidence in decisions. Decisions made from an ML and AI application combined with advanced optimization algorithms will improve the mission success and performance of systems, which will improve the overall cost and effectiveness of any program. The Systecon team constructs and employs model-based simulations, cutting across traditional silos of data, aggregating maintenance, and supply data, incorporating sensor information, and applying optimization and simulation methods to an as-maintained digital twin with the ability to aggregate results across a system’s lifecycle and across logical and operational groupings of systems. This coupling of data throughout the enterprise enables tactical, operational, and strategic decision support, detachable and deployable logistics services, and configuration-based automated distribution of digital technical and product data to enhance supply and logistics operations. As a complete solution, this approach significantly reduces program risk by allowing flexible configuration of data, data relationships, business process workflows, and early test and evaluation, especially budget trade-off analyses. A true capability to tie resources (dollars) to weapon system readiness in alignment with the real-world scenarios a warfighter may experience has been an objective yet to be realized to date. By developing and solidifying an organic capability to directly relate dollars to readiness and to inform the digital twin, the decision-maker is now empowered through valuable insight and traceability. This type of educated decision-making provides an advantage over the adversaries who struggle with maintaining system readiness at an affordable cost. The M&S capability developed allows program managers to independently evaluate system design and support decisions by quantifying their impact on operational availability and operations and support cost resulting in the ability to simultaneously optimize readiness and cost. This will allow the stakeholders to make data-driven decisions when trading cost and readiness throughout the life of the program. Finally, sponsors are available to validate product deliverables with efficiency and much higher accuracy than in previous years.

Keywords: artificial intelligence, digital transformation, machine learning, predictive analytics

Procedia PDF Downloads 160

Authors: Ismaila Olabisi Ogundiji, Hakeem Mayowa Olujide, Qasim Usamot

Abstract:

The main reasons for site-specific management for agricultural inputs are to increase the profitability of crop production, to protect the environment and to improve products’ quality. Information about the variability of different soil attributes within a field is highly essential for the decision-making process. Lack of fast and accurate acquisition of soil characteristics remains one of the biggest limitations of precision agriculture due to being expensive and time-consuming. Adaptive sampling has been proven as an accurate and affordable sampling technique for planning within a field for site-specific management of agricultural inputs. This study employed spatial uncertainty of soil apparent electrical conductivity (ECa) estimates to identify adaptive re-survey areas in the field. The original dataset was grouped into validation and calibration groups where the calibration group was sub-grouped into three sets of different measurements pass intervals. A conditional simulation was performed on the field ECa to evaluate the ECa spatial uncertainty estimates by the use of the geostatistical technique. The grouping of high-uncertainty areas for each set was done using image segmentation in MATLAB, then, high and low area value-separate was identified. Finally, an adaptive re-survey was carried out on those areas of high-uncertainty. Adding adaptive re-surveying significantly minimized the time required for resampling whole field and resulted in ECa with minimal error. For the most spacious transect, the root mean square error (RMSE) yielded from an initial crude sampling survey was minimized after an adaptive re-survey, which was close to that value of the ECa yielded with an all-field re-survey. The estimated sampling time for the adaptive re-survey was found to be 45% lesser than that of all-field re-survey. The results indicate that designing adaptive sampling through spatial uncertainty models significantly mitigates sampling cost, and there was still conformity in the accuracy of the observations.

Keywords: soil electrical conductivity, adaptive sampling, conditional simulation, spatial uncertainty, site-specific management

Procedia PDF Downloads 132

Authors: Sanjeev Prasher, T. Sai Vijay, Chandan Parsad, Abhishek Banerjee, Sahakari Nikhil Krishna, Subham Chatterjee

Abstract:

With rapid rise in internet retailing, the industry is set for a major implosion. Due to the little difference among competitors, companies find it difficult to segment and target the right shoppers. The objective of the study is to segment Indian online shoppers on the basis of the factors – website characteristics and shopping values. Together, these cover extrinsic and intrinsic factors that affect shoppers as they visit web retailers. Data were collected using questionnaire from 319 Indian online shoppers, and factor analysis was used to confirm the factors influencing the shoppers in their selection of web portals. Thereafter, cluster analysis was applied, and different segments of shoppers were identified. The relationship between income groups and online shoppers’ segments was tracked using correspondence analysis. Significant findings from the study include that web entertainment and informativeness together contribute more than fifty percent of the total influence on the web shoppers. Contrary to general perception that shoppers seek utilitarian leverages, the present study highlights the preference for fun, excitement, and entertainment during browsing of the website. Four segments namely Information Seekers, Utility Seekers, Value Seekers and Core Shoppers were identified and profiled. Value seekers emerged to be the most dominant segment with two-fifth of the respondents falling for hedonic as well as utilitarian shopping values. With overlap among the segments, utilitarian shopping value garnered prominence with more than fifty-eight percent of the total respondents. Moreover, a strong relation has been established between the income levels and the segments of Indian online shoppers. Web shoppers show different motives from being utility seekers to information seekers, core shoppers and finally value seekers as income levels increase. Companies can strategically use this information for target marketing and align their web portals accordingly. This study can further be used to develop models revolving around satisfaction, trust and customer loyalty.

Keywords: online shopping, shopping values, effectiveness of information content, web informativeness, web entertainment, information seekers, utility seekers, value seekers, core shoppers

Procedia PDF Downloads 195

Authors: Joy Cao, Min Zhou

Abstract:

Purpose: Acute Type A aortic dissection is a well-known cause of extremely high mortality rate. A highly accurate and cost-effective non-invasive predictor is critically needed so that the patient can be treated at earlier stage. Although various CFD approaches have been tried to establish some prediction frameworks, they are sensitive to uncertainty in both image segmentation and boundary conditions. Tedious pre-processing and demanding calibration procedures requirement further compound the issue, thus hampering their clinical applicability. Using the latest physics informed deep learning methods to establish an accurate and cost-effective predictor framework are amongst the main goals for a better Type A aortic dissection treatment. Methods: Via training a novel physics-informed deep residual network, with non-invasive 4D MRI displacement vectors as inputs, the trained model can cost-effectively calculate all these biomarkers: aortic blood pressure, WSS, and OSI, which are used to predict potential type A aortic dissection to avoid the high mortality events down the road. Results: The proposed deep learning method has been successfully trained and tested with both synthetic 3D aneurysm dataset and a clinical dataset in the aortic dissection context using Google colab environment. In both cases, the model has generated aortic blood pressure, WSS, and OSI results matching the expected patient’s health status. Conclusion: The proposed novel physics-informed deep residual network shows great potential to create a cost-effective, non-invasive predictor framework. Additional physics-based de-noising algorithm will be added to make the model more robust to clinical data noises. Further studies will be conducted in collaboration with big institutions such as Cleveland Clinic with more clinical samples to further improve the model’s clinical applicability.

Keywords: type-a aortic dissection, deep residual networks, blood flow modeling, data-driven modeling, non-invasive diagnostics, deep learning, artificial intelligence.

Procedia PDF Downloads 89

Authors: Berkan Ural, Arif Eser, Sinan Apaydin

Abstract:

Medical image processing is generally become a challenging task nowadays. Indeed, processing of brain MRI images is one of the difficult parts of this area. This study proposes a hybrid well-defined approach which is consisted from tumor detection, extraction and analyzing steps. This approach is mainly consisted from a computer aided diagnostics system for identifying and detecting the tumor formation in any region of the brain and this system is commonly used for early prediction of brain tumor using advanced image processing and probabilistic neural network methods, respectively. For this approach, generally, some advanced noise removal functions, image processing methods such as automatic segmentation and morphological operations are used to detect the brain tumor boundaries and to obtain the important feature parameters of the tumor region. All stages of the approach are done specifically with using MATLAB software. Generally, for this approach, firstly tumor is successfully detected and the tumor area is contoured with a specific colored circle by the computer aided diagnostics program. Then, the tumor is segmented and some morphological processes are achieved to increase the visibility of the tumor area. Moreover, while this process continues, the tumor area and important shape based features are also calculated. Finally, with using the probabilistic neural network method and with using some advanced classification steps, tumor area and the type of the tumor are clearly obtained. Also, the future aim of this study is to detect the severity of lesions through classes of brain tumor which is achieved through advanced multi classification and neural network stages and creating a user friendly environment using GUI in MATLAB. In the experimental part of the study, generally, 100 images are used to train the diagnostics system and 100 out of sample images are also used to test and to check the whole results. The preliminary results demonstrate the high classification accuracy for the neural network structure. Finally, according to the results, this situation also motivates us to extend this framework to detect and localize the tumors in the other organs.

Keywords: image processing algorithms, magnetic resonance imaging, neural network, pattern recognition

Procedia PDF Downloads 418

Authors: Christina Hagen, Pragathi Kamale Gurmurthy, Thorsten M. Buzug

Abstract:

CFD simulations are performed in the upper airway of a patient suffering from obstructive sleep apnea (OSA) that is a sleep related breathing disorder characterized by repetitive partial or complete closures of the upper airways. The simulations are aimed at getting a better understanding of the pathophysiological flow patterns in an OSA patient. The simulation is compared to medical data of a sleep endoscopic examination under sedation. A digital model consisting of surface triangles of the upper airway is extracted from the MR images by a region growing segmentation process and is followed by a careful manual refinement. The computational domain includes the nasal cavity with the nostrils as the inlet areas and the pharyngeal volume with an outlet underneath the larynx. At the nostrils a flat inflow velocity profile is prescribed by choosing the velocity such that a volume flow rate of 150 ml/s is reached. Behind the larynx at the outlet a pressure of -10 Pa is prescribed. The stationary incompressible Navier-Stokes equations are numerically solved using finite elements. A grid convergence study has been performed. The results show an amplification of the maximal velocity of about 2.5 times the inlet velocity at a constriction of the pharyngeal volume in the area of the tongue. It is the same region that also shows the highest pressure drop from about 5 Pa. This is in agreement with the sleep endoscopic examinations of the same patient under sedation showing complete contractions in the area of the tongue. CFD simulations can become a useful tool in the diagnosis and therapy of obstructive sleep apnea by giving insight into the patient’s individual fluid dynamical situation in the upper airways giving a better understanding of the disease where experimental measurements are not feasible. Within this study, it could been shown on one hand that constriction areas within the upper airway lead to a significant pressure drop and on the other hand a good agreement of the area of pressure drop and the area of contraction could be shown.

Keywords: biomedical engineering, obstructive sleep apnea, pharynx, upper airways

Procedia PDF Downloads 306

Authors: Swati Swati, Yuhang Chen, Robert Reuben

Abstract:

Coronary stents are devices resembling the shape of a tube which are placed in coronary arteries, to keep the arteries open in the treatment of coronary arterial diseases. Coronary stents are routinely deployed to clear atheromatous plaque. The stent essentially applies an internal pressure to the artery because its structure is cylindrically symmetrical and this may introduce some abnormalities in final arterial shape. The goal of the project is to develop segmented circumferential arterial compliance measuring devices which can be deployed (eventually) in vivo. The segmentation of the device will allow the mechanical asymmetry of any stenosis to be assessed. The purpose will be to assess the quality of arterial tissue for applications in tailored stents and in the assessment of aortic aneurism. Arterial distensibility measurement is of utmost importance to diagnose cardiovascular diseases and for prediction of future cardiac events or coronary artery diseases. In order to arrive at some generic outcomes, a preliminary experimental set-up has been devised to establish the measurement principles for the device at macro-scale. The measurement methodology consists of a strain gauge system monitored by LABVIEW software in a real-time fashion. This virtual instrument employs a balloon within a gelatine model contained in a split cylinder with strain gauges fixed on it. The instrument allows automated measurement of the effect of air-pressure on gelatine and measurement of strain with respect to time and pressure during inflation. Compliance simple creep model has been applied to the results for the purpose of extracting some measures of arterial compliance. The results obtained from the experiments have been used to study the effect of air pressure on strain at varying time intervals. The results clearly demonstrate that with decrease in arterial volume and increase in arterial pressure, arterial strain increases thereby decreasing the arterial compliance. The measurement system could lead to development of portable, inexpensive and small equipment and could prove to be an efficient automated compliance measurement device.

Keywords: arterial compliance, atheromatous plaque, mechanical symmetry, strain measurement

Procedia PDF Downloads 279

Authors: Manpreet Kaur, Cheol-Hong Min

Abstract:

The increasing advancement in the robotics equipped with machine vision sensors applied to precision agriculture is a demanding solution for various problems in the agricultural farms. An important issue related with the machine vision system concerns crop row and weed detection. This paper proposes an automatic furrow detection system based on real-time processing for identifying crop rows in maize fields in the presence of weed. This vision system is designed to be installed on the farming vehicles, that is, submitted to gyros, vibration and other undesired movements. The images are captured under image perspective, being affected by above undesired effects. The goal is to identify crop rows for vehicle navigation which includes weed removal, where weeds are identified as plants outside the crop rows. The images quality is affected by different lighting conditions and gaps along the crop rows due to lack of germination and wrong plantation. The proposed image processing method consists of four different processes. First, image segmentation based on HSV (Hue, Saturation, Value) decision tree. The proposed algorithm used HSV color space to discriminate crops, weeds and soil. The region of interest is defined by filtering each of the HSV channels between maximum and minimum threshold values. Then the noises in the images were eliminated by the means of hybrid median filter. Further, mathematical morphological processes, i.e., erosion to remove smaller objects followed by dilation to gradually enlarge the boundaries of regions of foreground pixels was applied. It enhances the image contrast. To accurately detect the position of crop rows, the region of interest is defined by creating a binary mask. The edge detection and Hough transform were applied to detect lines represented in polar coordinates and furrow directions as accumulations on the angle axis in the Hough space. The experimental results show that the method is effective.

Keywords: furrow detection, morphological, HSV, Hough transform

Procedia PDF Downloads 231

Authors: Filip Rusak, Yulia Arzhaeva, Dadong Wang

Abstract:

Machine learning based Computer-Aided Diagnosis (CAD) is gaining much popularity in medical imaging and diagnostic radiology. However, it requires a large amount of high quality and labeled training image datasets. The training images may come from different sources and be acquired from different radiography machines produced by different manufacturers, digital or digitized copies of film radiographs, with various sizes as well as different pixel intensity distributions. In this paper, a content-aware image augmentation method is presented to deal with these variations. The results of the proposed method have been validated graphically by plotting the removed and added seams of pixels on original images. Two different chest X-ray (CXR) datasets are used in the experiments. The CXRs in the datasets defer in size, some are digital CXRs while the others are digitized from analog CXR films. With the proposed content-aware augmentation method, the Seam Carving algorithm is employed to resize CXRs and the corresponding labels in the form of image masks, followed by histogram matching used to normalize the pixel intensities of digital radiography, based on the pixel intensity values of digitized radiographs. We implemented the algorithms, resized the well-known Montgomery dataset, to the size of the most frequently used Japanese Society of Radiological Technology (JSRT) dataset and normalized our digital CXRs for testing. This work resulted in the unified off-the-shelf CXR dataset composed of radiographs included in both, Montgomery and JSRT datasets. The experimental results show that even though the amount of augmentation is large, our algorithm can preserve the important information in lung fields, local structures, and global visual effect adequately. The proposed method can be used to augment training and testing image data sets so that the trained machine learning model can be used to process CXRs from various sources, and it can be potentially used broadly in any medical imaging applications.

Keywords: computer-aided diagnosis, image augmentation, lung segmentation, medical imaging, seam carving

Procedia PDF Downloads 222