Search results for: convolutional coding

Authors: Emily Kambalame

Abstract:

Water distribution networks necessitate many investments for construction, prompting researchers to seek cost reduction and efficient design solutions. Optimization techniques are employed in this regard to address these challenges. In this context, the penalty-free multi-objective evolutionary algorithm (PFMOEA) coupled with pressure-dependent analysis (PDA) was utilized to develop a multi-objective evolutionary search for the optimization of water distribution systems (WDSs). The aim of this research was to find out if the computational efficiency of the PFMOEA for WDS optimization could be enhanced. This was done by applying real coding representation and retaining different percentages of feasible and infeasible solutions close to the Pareto front in the elitism step of the optimization. Two benchmark network problems, namely the Two-looped and Hanoi networks, were utilized in the study. A comparative analysis was then conducted to assess the performance of the real-coded PFMOEA in relation to other approaches described in the literature. The algorithm demonstrated competitive performance for the two benchmark networks by implementing real coding. The real-coded PFMOEA achieved the novel best-known solutions ($419,000 and $6.081 million) and a zero-pressure deficit for the two networks, requiring fewer function evaluations than the binary-coded PFMOEA. In previous PFMOEA studies, elitism applied a default retention of 30% of the least cost-feasible solutions while excluding all infeasible solutions. It was found in this study that by replacing 10% and 15% of the feasible solutions with infeasible ones that are close to the Pareto front with minimal pressure deficit violations, the computational efficiency of the PFMOEA was significantly enhanced. The configuration of 15% feasible and 15% infeasible solutions outperformed other retention allocations by identifying the optimal solution with the fewest function evaluation

Keywords: design optimization, multi-objective evolutionary, penalty-free, water distribution systems

Procedia PDF Downloads 62

Authors: Mrinal Kanti Bhowmik, Debanjana Debnath Jr., Debotosh Bhattacharjee

Abstract:

A facial expression is undeniably the human manners. It is a significant channel for human communication and can be applied to extract emotional features accurately. People in pain often show variations in facial expressions that are readily observable to others. A core of actions is likely to occur or to increase in intensity when people are in pain. To illustrate the changes in the facial appearance, a system known as Facial Action Coding System (FACS) is pioneered by Ekman and Friesen for human observers. According to Prkachin and Solomon, a set of such actions carries the bulk of information about pain. Thus, the Prkachin and Solomon pain intensity (PSPI) metric is defined. So, it is very important to notice that facial expressions, being a behavioral source in communication media, provide an important opening into the issues of non-verbal communication in pain. People express their pain in many ways, and this pain behavior is the basis on which most inferences about pain are drawn in clinical and research settings. Hence, to understand the roles of different pain behaviors, it is essential to study the properties. For the past several years, the studies are concentrated on the properties of one specific form of pain behavior i.e. facial expression. This paper represents a comprehensive study on pain assessment that can model and estimate the intensity of pain that the patient is suffering. It also reviews the historical background of different pain assessment techniques in the context of painful expressions. Different approaches incorporate FACS from psychological views and a pain intensity score using the PSPI metric in pain estimation. This paper investigates in depth analysis of different approaches used in pain estimation and presents different observations found from each technique. It also offers a brief study on different distinguishing features of real and fake pain. Therefore, the necessity of the study lies in the emerging fields of painful face assessment in clinical settings.

Keywords: facial action coding system (FACS), pain, pain behavior, Prkachin and Solomon pain intensity (PSPI)

Procedia PDF Downloads 346

Authors: Steven J. Singer, Julianna F. Kamenakis, Allison R. Shapiro, Kimberly M. Cacciato

Abstract:

Communication and language are topics frequently studied among deaf children. However, there is limited research that focuses specifically on the communication and language experiences of Deaf-Disabled children. In this ethnography, researchers investigated the language experiences of six sets of parents with Deaf-Disabled children who chose American Sign Language (ASL) as the preferred mode of communication for their child. Specifically, the researchers were interested in the factors that influenced the parents’ decisions regarding their child’s communication options, educational placements, and social experiences. Data collection in this research included 18 hours of semi-structured interviews, 20 hours of participant observations, over 150 pages of reflexive journals and field notes, and a 2-hour focus group. The team conducted constant comparison qualitative analysis using NVivo software and an inductive coding procedure. The four researchers each read the data several times until they were able to chunk it into broad categories about communication and social influences. The team compared the various categories they developed, selecting ones that were consistent among researchers and redefining categories that differed. Continuing to use open inductive coding, the research team refined the categories until they were able to develop distinct themes. Two team members developed each theme through a process of independent coding, comparison, discussion, and resolution. The research team developed three themes: 1) early medical needs provided time for the parents to explore various communication options for their Deaf-Disabled child, 2) without intervention from medical professionals or educators, ASL emerged as a prioritized mode of communication for the family, 3) atypical gender roles affected familial communication dynamics. While managing the significant health issues of their Deaf-Disabled child at birth, families and medical professionals were so fixated on tending to the medical needs of the child that the typical pressures of determining a mode of communication were deprioritized. This allowed the families to meticulously research various methods of communication, resulting in an informed, rational, and well-considered decision to use ASL as the primary mode of communication with their Deaf-Disabled child. It was evident that having a Deaf-Disabled child meant an increased amount of labor and responsibilities for parents. This led to a shift in the roles of the family members. During the child’s development, the mother transformed from fulfilling the stereotypical roles of nurturer and administrator to that of administrator and champion. The mother facilitated medical proceedings and educational arrangements while the father became the caretaker and nurturer of their Deaf-Disabled child in addition to the traditional role of earning the family’s primary income. Ultimately, this research led to a deeper understanding of the critical role that time plays in parents’ decision-making process regarding communication methods with their Deaf-Disabled child.

Keywords: American Sign Language, deaf-disabled, ethnography, sociolinguistics

Procedia PDF Downloads 120

Authors: Maria Kountouridou, Dino Domic

Abstract:

Within an increasingly competitive and dynamic environment, the higher education sector is becoming more commodified, with the concept of branding to become exceedingly imperative and an inextricable ingredient for the university’s success. Branding in higher education has proven to be an effective strategy that managed to receive considerable attention in the recent few years, and a growing number of articles have begun to appear in the literature. However, a clear void in the literature confirms that the concept of students’ perceptions towards the university’s brand image has not been researched extensively. An investigation on this central concept is of paramount importance since it will facilitate the development of an inductively generated theoretical model concerning branding in higher education. This research focuses on examining the impact of the ‘positive-visualization-course in brand identity’ upon the perception of freshmen students towards a university’s brand image. A grounded theory methodology has been selected, consisting of semi-structured interviews. Forty-two students have participated in the research, among which twenty-five women and seventeen men. The identification of the sample emerged through the use of the snowball sampling technique. The participants were divided into two groups (experimental and control group) after the researcher had taken into consideration the factor ‘program of study’, to eliminate any possible interaction between the participants of each group. An experiment was carried out where a ‘positive-visualization-course in brand identity’ was conducted among the participants of the experimental group, while the participants of the control group have not been exposed to the course. For the purpose of this research, the term ‘positive-visualization-course in brand identity’ refers to a course where brand history, past achievements/recognitions/awards, its values, and its mission are presented. Prior to the course implementation, face-to-face semi-structured interviews were carried out among the participants of both groups, with the aim of examining the freshmen students’ perceptions towards the university’s brand image. One week after the course implementation, the researcher carried out semi-structured interviews with the participants of the experimental group only in order to identify whether students’ perceptions had been affected after the course completion. Four months after the course completion, semi-structured interviews were carried out among the participants of both groups. Eight months after the course completion, semi-structured interviews were conducted with the aim of identifying the freshmen students’ updated perceptions. Data has been analyzed using substantive coding (open and selective coding), theoretical coding, field memos, and constant comparative analysis. The findings strongly suggest that the ‘positive-visualization-course in brand identity’ can positively affect freshmen students’ perceptions towards a university’s brand image. Additionally, other factors conduce to the formation of perception throughout the months. This study contributes and expands upon the existing literature by presenting an inductively generated theoretical model to guide future research in the links between ‘positive-visualization-course in brand identity’ and the perception of freshmen students towards a university’s brand image.

Keywords: brand image, brand name, branding, higher education marketing, perception

Procedia PDF Downloads 178

Authors: Mustafa El-Halabi

Abstract:

A general model of wiretap channel with states is considered, where the legitimate receiver and the wiretapper’s observations depend on three states S1, S2 and S3. State S1 is non-causally known to the encoder, S2 is known to the receiver, and S3 remains unknown. A secure coding scheme, based using structured-binning, is proposed, and it is shown to achieve the secrecy capacity when the signal at legitimate receiver is a deterministic function of the input.

Keywords: physical layer security, interference, side information, secrecy capacity

Procedia PDF Downloads 389

Authors: Masoomeh Alsadat Mirshafaei

Abstract:

The present study aims to explore the effect of an 8-week aerobic training regimen combined with cinnamon extract on serum irisin and leptin levels in insulin-resistant rats. Additionally, this research leverages various machine learning (ML) and deep learning (DL) algorithms to model the complex interdependencies between exercise, nutrition, and metabolic markers, offering a groundbreaking approach to obesity and diabetes research. Forty-eight Wistar rats were selected and randomly divided into four groups: control, training, cinnamon, and training cinnamon. The training protocol was conducted over 8 weeks, with sessions 5 days a week at 75-80% VO2 max. The cinnamon and training-cinnamon groups were injected with 200 ml/kg/day of cinnamon extract. Data analysis included serum data, dietary intake, exercise intensity, and metabolic response variables, with blood samples collected 72 hours after the final training session. The dataset was analyzed using one-way ANOVA (P<0.05) and fed into various ML and DL models, including Support Vector Machines (SVM), Random Forest (RF), and Convolutional Neural Networks (CNN). Traditional statistical methods indicated that aerobic training, with and without cinnamon extract, significantly increased serum irisin and decreased leptin levels. Among the algorithms, the CNN model provided superior performance in identifying specific interactions between cinnamon extract concentration and exercise intensity, optimizing the increase in irisin and the decrease in leptin. The CNN model achieved an accuracy of 92%, outperforming the SVM (85%) and RF (88%) models in predicting the optimal conditions for metabolic marker improvements. The study demonstrated that advanced ML and DL techniques could uncover nuanced relationships and potential cellular responses to exercise and dietary supplements, which is not evident through traditional methods. These findings advocate for the integration of advanced analytical techniques in nutritional science and exercise physiology, paving the way for personalized health interventions in managing obesity and diabetes.

Keywords: aerobic training, cinnamon extract, insulin resistance, irisin, leptin, convolutional neural networks, exercise physiology, support vector machines, random forest

Procedia PDF Downloads 38

Authors: Abhaydeep Pandey, Shweta Shukla, Saptamita Goswami, Bhaswati Bandyopadhyay, Vishnampettai Ramachandran, Sudhanshu Vrati, Arup Banerjee

Abstract:

Dengue virus infection is now considered as one of the most important mosquito-borne infection in human. The virus is known to promote vascular permeability, cerebral edema leading to Dengue hemorrhagic fever (DHF) or Dengue shock syndrome (DSS). Dengue infection has known to be endemic in India for over two centuries as a benign and self-limited disease. In the last couple of years, the disease symptoms have changed, manifesting severe secondary complication. So far, Delhi has experienced 12 outbreaks of dengue virus infection since 1997 with the last reported in 2014-15. Without specific antivirals, the case management of high-risk dengue patients entirely relies on supportive care, involving constant monitoring and timely fluid support to prevent hypovolemic shock. Nonetheless, the diverse clinical spectrum of dengue disease, as well as its initial similarity to other viral febrile illnesses, presents a challenge in the early identification of this high-risk group. WHO recommends the use of warning signs to identify high-risk patients, but warning signs generally appear during, or just one day before the development of severe illness, thus, providing only a narrow window for clinical intervention. The ability to predict which patient may develop DHF and DSS may improve the triage and treatment. With the recent discovery of high throughput RNA sequencing allows us to understand the disease progression at the genomic level. Here, we will collate the results of RNA-Sequencing data obtained recently from PBMC of different categories of dengue patients from India and will discuss the possible role of deregulated genes and long non-coding RNAs NEAT1 for development of disease progression.

Keywords: long non-coding RNA (lncRNA), dengue, peripheral blood mononuclear cell (PBMC), nuclear enriched abundant transcript 1 (NEAT1), dengue hemorrhagic fever (DHF), dengue shock syndrome (DSS)

Procedia PDF Downloads 308

Authors: Guezouli Larbi, Belkacem Soundes

Abstract:

The automation of the text detection process can help the human in his driving task. Its application can be very useful to help drivers to have more information about their environment by facilitating the reading of road signs such as directional signs, events, stores, etc. In this paper, a system consisting of two stages has been proposed. In the first one, we used pseudo-Zernike moments to pinpoint areas of the image that may contain text. The architecture of this part is based on three main steps, region of interest (ROI) detection, text localization, and non-text region filtering. Then, in the second step, we present a convolutional neural network architecture (On-Road Text Detection Network - ORTDN) which is considered a classification phase. The results show that the proposed framework achieved ≈ 35 fps and an mAP of ≈ 90%, thus a low computational time with competitive accuracy.

Keywords: text detection, CNN, PZM, deep learning

Procedia PDF Downloads 83

Authors: Vedant Vardhan Rathour, Anant Agrawal

Abstract:

The project aims to provide a touchless, intuitive interface for human-computer interaction, enabling users to control their computer using hand gestures and voice commands. The system leverages advanced computer vision techniques using the MediaPipe framework and OpenCV to detect and interpret real time hand gestures, transforming them into mouse actions such as clicking, dragging, and scrolling. Additionally, the integration of a voice assistant powered by the Speech Recognition library allows for seamless execution of tasks like web searches, location navigation and gesture control on the system through voice commands.

Keywords: gesture recognition, hand tracking, machine learning, convolutional neural networks

Procedia PDF Downloads 12

Authors: Adel Edwar Waheeb Louka

Abstract:

Purpose AI-driven solutions are at the forefront of many pathology and medical imaging methods. Using algorithms designed to better the experience of medical professionals within their respective fields, the efficiency and accuracy of diagnosis can improve. In particular, X-rays are a fast and relatively inexpensive test that can diagnose diseases. In recent years, X-rays have not been widely used to detect and diagnose COVID-19. The under use of Xrays is mainly due to the low diagnostic accuracy and confounding with pneumonia, another respiratory disease. However, research in this field has expressed a possibility that artificial neural networks can successfully diagnose COVID-19 with high accuracy. Models and Data The dataset used is the COVID-19 Radiography Database. This dataset includes images and masks of chest X-rays under the labels of COVID-19, normal, and pneumonia. The classification model developed uses an autoencoder and a pre-trained convolutional neural network (DenseNet201) to provide transfer learning to the model. The model then uses a deep neural network to finalize the feature extraction and predict the diagnosis for the input image. This model was trained on 4035 images and validated on 807 separate images from the ones used for training. The images used to train the classification model include an important feature: the pictures are cropped beforehand to eliminate distractions when training the model. The image segmentation model uses an improved U-Net architecture. This model is used to extract the lung mask from the chest X-ray image. The model is trained on 8577 images and validated on a validation split of 20%. These models are calculated using the external dataset for validation. The models’ accuracy, precision, recall, f1-score, IOU, and loss are calculated. Results The classification model achieved an accuracy of 97.65% and a loss of 0.1234 when differentiating COVID19-infected, pneumonia-infected, and normal lung X-rays. The segmentation model achieved an accuracy of 97.31% and an IOU of 0.928. Conclusion The models proposed can detect COVID-19, pneumonia, and normal lungs with high accuracy and derive the lung mask from a chest X-ray with similarly high accuracy. The hope is for these models to elevate the experience of medical professionals and provide insight into the future of the methods used.

Keywords: artificial intelligence, convolutional neural networks, deeplearning, image processing, machine learningSarapin, intraarticular, chronic knee pain, osteoarthritisFNS, trauma, hip, neck femur fracture, minimally invasive surgery

Procedia PDF Downloads 73

Authors: Denis Mustafov, Emmanouil Karteris, Maria Braoudaki

Abstract:

Glioblastoma multiform (GBM) is a heterogenous primary brain tumour that kills most affected patients. To the authors best knowledge, despite all research efforts there is no early diagnostic biomarker for GBM. MicroRNAs (miRNAs) are short non-coding RNA molecules which are deregulated in many cancers. The aim of this research was to determine miRNAs with a diagnostic impact and to potentially identify promising therapeutic targets for glioblastoma multiform. In silico analysis was performed to identify deregulated miRNAs with diagnostic relevance for glioblastoma. The expression profiles of the chosen miRNAs were then validated in vitro in the human glioblastoma cell lines A172 and U-87MG. Briefly, RNA extraction was carried out using the Trizol method, whilst miRNA extraction was performed using the mirVANA miRNA isolation kit. Quantitative Real-Time Polymerase Chain Reaction was performed to verify their expression. The presence of five target proteins within the A172 cell line was evaluated by Western blotting. The expression of the CORO1C protein within 32 GBM cases was examined via immunohistochemistry. The miRNAs identified in silico included miR-21-5p, miR-34a and miR-128a. These miRNAs were shown to target deregulated GBM genes, such as CDK6, E2F3, BMI1, JAG1, and CORO1C. miR-34a and miR-128a showed low expression profiles in comparison to a control miR-RNU-44 in both GBM cell lines suggesting tumour suppressor properties. Opposing, miR-21-5p demonstrated greater expression indicating that it could potentially function as an oncomiR. Western blotting revealed expression of all five proteins within the A172 cell line. In silico analysis also suggested that CORO1C is a target of miR-128a and miR-34a. Immunohistochemistry demonstrated that 75% of the GBM cases showed moderate to high expression of CORO1C protein. Greater understanding of the deregulated expression of miR-128a and the upregulation of CORO1C in GBM could potentially lead to the identification of a promising diagnostic biomarker signature for glioblastomas.

Keywords: non-coding RNAs, gene expression, brain tumours, immunohistochemistry

Procedia PDF Downloads 89

Authors: Catia Correia Caeiro, Kun Guo, Daniel Mills

Abstract:

Despite its growing interest, emotions are still an understudied cognitive process and their origins are currently the focus of much debate among the scientific community. The use of facial expressions as traditional hallmarks of discrete and holistic emotions created a circular reasoning due to a priori assumptions of meaning and its associated appearance-biases. Ekman and colleagues solved this problem and laid the foundations for the quantitative and systematic study of facial expressions in humans by developing an anatomically-based system (independent from meaning) to measure facial behaviour, the Facial Action Coding System (FACS). One way of investigating emotion cognition processes is by applying comparative psychology methodologies and looking at either closely-related species (e.g. chimpanzees) or phylogenetically distant species sharing similar present adaptation problems (analogy). In this study, the domestic dog was used as a comparative animal model to look at facial expressions in social interactions in parallel with human facial expressions. The orofacial musculature seems to be relatively well conserved across mammal species and the same holds true for the domestic dog. Furthermore, the dog is unique in having shared the same social environment as humans for more than 10,000 years, facing similar challenges and acquiring a unique set of socio-cognitive skills in the process. In this study, the spontaneous facial movements of humans and dogs were compared when interacting with hetero- and conspecifics as well as in solitary contexts. In total, 200 participants were examined with FACS and DogFACS (The Dog Facial Action Coding System): coding tools across four different emotionally-driven contexts: a) Happiness (play and reunion), b) anticipation (of positive reward), c) fear (object or situation triggered), and d) frustration (negation of a resource). A neutral control was added for both species. All four contexts are commonly encountered by humans and dogs, are comparable between species and seem to give rise to emotions from homologous brain systems. The videos used in the study were extracted from public databases (e.g. Youtube) or published scientific databases (e.g. AM-FED). The results obtained allowed us to delineate clear similarities and differences on the flexibility of the facial musculature in the two species. More importantly, they shed light on what common facial movements are a product of the emotion linked contexts (the ones appearing in both species) and which are characteristic of the species, revealing an important clue for the debate on the origin of emotions. Additionally, we were able to examine movements that might have emerged for interspecific communication. Finally, our results are discussed from an evolutionary perspective adding to the recent line of work that supports an ancient shared origin of emotions in a mammal ancestor and defining emotions as mechanisms with a clear adaptive purpose essential on numerous situations, ranging from maintenance of social bonds to fitness and survival modulators.

Keywords: comparative and evolutionary psychology, emotion, facial expressions, FACS

Procedia PDF Downloads 434

Authors: Ross Goroshin, Joan Bruna, Jonathan Tompson, David Eigen, Yann LeCun

Abstract:

Current state-of-the-art classification and detection algorithms rely on supervised training. In this work we study unsupervised feature learning in the context of temporally coherent video data. We focus on feature learning from unlabeled video data, using the assumption that adjacent video frames contain semantically similar information. This assumption is exploited to train a convolutional pooling auto-encoder regularized by slowness and sparsity. We establish a connection between slow feature learning to metric learning and show that the trained encoder can be used to define a more temporally and semantically coherent metric.

Keywords: machine learning, pattern clustering, pooling, classification

Procedia PDF Downloads 456

Authors: Tyler T. Procko, Steve Collins

Abstract:

New features in .NET (6 and above) permit streamlined access to information residing in JSON-capable relational databases, such as SQL Server (2016 and above). Traditional methods of data access now comparatively involve unnecessary steps which compromise system performance. This work posits that the established ORM (Object Relational Mapping) based methods of data access in applications and APIs result in common issues, e.g., object-relational impedance mismatch. Recent developments in C# and .NET Core combined with a framework of modern SQL Server coding conventions have allowed better technical solutions to the problem. As an amelioration, this work details the language features and coding conventions which enable this streamlined approach, resulting in an open-source .NET library implementation called Codeless Data Access (CODA). Canonical approaches rely on ad-hoc mapping code to perform type conversions between the client and back-end database; with CODA, no mapping code is needed, as JSON is freely mapped to SQL and vice versa. CODA streamlines API data access by improving on three aspects of immediate concern to web developers, database engineers and cybersecurity professionals: Simplicity, Speed and Security. Simplicity is engendered by cutting out the “middleman” steps, effectively making API data access a whitebox, whereas traditional methods are blackbox. Speed is improved because of the fewer translational steps taken, and security is improved as attack surfaces are minimized. An empirical evaluation of the speed of the CODA approach in comparison to ORM approaches ] is provided and demonstrates that the CODA approach is significantly faster. CODA presents substantial benefits for API developer workflows by simplifying data access, resulting in better speed and security and allowing developers to focus on productive development rather than being mired in data access code. Future considerations include a generalization of the CODA method and extension outside of the .NET ecosystem to other programming languages.

Keywords: API data access, database, JSON, .NET core, SQL server

Procedia PDF Downloads 66

Authors: Ashutosh Mishra, Nikhil Goyal

Abstract:

A 3D facial emotion recognition model based on deep learning is proposed in this paper. Two convolution layers and a pooling layer are employed in the deep learning architecture. After the convolution process, the pooling is finished. The probabilities for various classes of human faces are calculated using the sigmoid activation function. To verify the efficiency of deep learning-based systems, a set of faces. The Kaggle dataset is used to verify the accuracy of a deep learning-based face recognition model. The model's accuracy is about 65 percent, which is lower than that of other facial expression recognition techniques. Despite significant gains in representation precision due to the nonlinearity of profound image representations.

Keywords: facial recognition, computational intelligence, convolutional neural network, depth map

Procedia PDF Downloads 231

Authors: Huayuan Lu, Chunfang Yang, Ma Zhu, Baojun Qi, Yaqiong Qiao, Jiangqian Xu

Abstract:

Image geolocalization has great application prospects in fields such as autonomous driving and virtual/augmented reality. In practical application scenarios, the size of the image to be located is not fixed; it is impractical to train different networks for all possible sizes. When its size does not match the size of the input of the descriptor extraction model, existing image geolocalization methods usually directly scale or crop the image in some common ways. This will result in the loss of some information important to the geolocalization task, thus affecting the performance of the image geolocalization method. For example, excessive down-sampling can lead to blurred building contour, and inappropriate cropping can lead to the loss of key semantic elements, resulting in incorrect geolocation results. To address this problem, this paper designs a learnable image resizer and proposes an arbitrary-sized image geolocation method. (1) The designed learnable image resizer employs the self-attention mechanism to enhance the geometric features of the resized image. Firstly, it applies bilinear interpolation to the input image and its feature maps to obtain the initial resized image and the resized feature maps. Then, SKNet (selective kernel net) is used to approximate the best receptive field, thus keeping the geometric shapes as the original image. And SENet (squeeze and extraction net) is used to automatically select the feature maps with strong contour information, enhancing the geometric features. Finally, the enhanced geometric features are fused with the initial resized image, to obtain the final resized images. (2) The proposed image geolocalization method embeds the above image resizer as a fronting layer of the descriptor extraction network. It not only enables the network to be compatible with arbitrary-sized input images but also enhances the geometric features that are crucial to the image geolocalization task. Moreover, the triplet attention mechanism is added after the first convolutional layer of the backbone network to optimize the utilization of geometric elements extracted by the first convolutional layer. Finally, the local features extracted by the backbone network are aggregated to form image descriptors for image geolocalization. The proposed method was evaluated on several mainstream datasets, such as Pittsburgh30K, Tokyo24/7, and Places365. The results show that the proposed method has excellent size compatibility and compares favorably to recently mainstream geolocalization methods.

Keywords: image geolocalization, self-attention mechanism, image resizer, geometric feature

Procedia PDF Downloads 214

Authors: Vidhi Jain

Abstract:

Communication between speakers and non-speakers of American Sign Language (ASL) can be problematic, inconvenient, and expensive. This project attempts to bridge the communication gap by designing a portable glove that captures the user’s ASL gestures and outputs the translated text on a smartphone. The glove is equipped with flex sensors, contact sensors, and a gyroscope to measure the flexion of the fingers, the contact between fingers, and the rotation of the hand. The glove’s Arduino UNO microcontroller analyzes the sensor readings to identify the gesture from a library of learned gestures. The Bluetooth module transmits the gesture to a smartphone. Using this device, one day speakers of ASL may be able to communicate with others in an affordable and convenient way.

Keywords: sign language, morse code, convolutional neural network, American sign language, gesture recognition

Procedia PDF Downloads 63

Authors: Sushil Kumar Gupta, Atharva Joshi, Ayush Sonawale, Sachin Naik, Rajshree Khande

Abstract:

The usage of artificially created videos and audio by deep learning is a major problem of the current media landscape, as it pursues the goal of misinformation and distrust. In conclusion, the objective of this work targets generating a reliable deepfake detection model using deep learning that will help detect forged videos accurately. In this work, CelebDF v1, one of the largest deepfake benchmark datasets in the literature, is adopted to train and test the proposed models. The data includes authentic and synthetic videos of high quality, therefore allowing an assessment of the model’s performance against realistic distortions.

Keywords: deepfake detection, CelebDF v1, convolutional neural network (CNN), xception model, data augmentation, media manipulation

Procedia PDF Downloads 10

Authors: Salman Akrokayan

Abstract:

Synthetic cDNA (ScDNA) of granulocyte colony-stimulating factor (G-CSF) was constructed using a DNA synthesizer with the aim to increase its expression level. 5' end of the ScDNA of G-CSF coding region was modified by decreasing the GC content without altering the predicted amino acids sequence. The identity of the resulting protein from ScDNA was confirmed by the highly specific enzyme-linked immunosorbent assay. In conclusion, a synthetic G-CSF cDNA in combination with the recombinant DNA protocol offers a rapid and reliable strategy for synthesizing the target protein. However, the commercial utilization of this methodology requires rigorous validation and quality control.

Keywords: synthetic cDNA, recombinant G-CSF, cloning, gene expression

Procedia PDF Downloads 285

Authors: C. Mayr, J. Periya, A. Kariminezhad

Abstract:

In this paper, we developed a learning framework for the classification of vulnerable road users (VRU) by their range-Doppler signatures. The frequency-modulated continuous-wave (FMCW) radar raw data is first pre-processed to obtain robust object range-Doppler maps per coherent time interval. The complex-valued range-Doppler maps captured from our outdoor measurements are further fed into a convolutional neural network (CNN) to learn the classification. This CNN has gone through a hyperparameter optimization process for improved learning. By learning VRU range-Doppler signatures, the three classes 'pedestrian', 'dog', and 'noise' are classified with an average accuracy of almost 95%. Interestingly, this classification accuracy holds for a combined longitudinal and lateral object trajectories.

Keywords: machine learning, radar, signal processing, autonomous driving

Procedia PDF Downloads 246

Authors: Dev Pratap Singh, Harshika Hasija, Ashwini S.

Abstract:

The project aims to provide a touchless, intuitive interface for human-computer interaction, enabling users to control their computers using hand gestures and voice commands. The system leverages advanced computer vision techniques using the Media Pipe framework and OpenCV to detect and interpret real-time hand gestures, transforming them into mouse actions such as clicking, dragging, and scrolling. Additionally, the integration of a voice assistant powered by the speech recognition library allows for seamless execution of tasks like web searches, location navigation, and gesture control in the system through voice commands.

Keywords: gesture recognition, hand tracking, machine learning, convolutional neural networks, natural language processing, voice assistant

Procedia PDF Downloads 10

Authors: Lin Cheng, Zijiang Yang

Abstract:

Program synthesis is the task to automatically generate programs based on user specification. In this paper, we present a framework that synthesizes programs from flow charts that serve as accurate and intuitive specification. In order doing so, we propose a deep neural network called GRCNN that recognizes graph structure from its image. GRCNN is trained end-to-end, which can predict edge and node information of the flow chart simultaneously. Experiments show that the accuracy rate to synthesize a program is 66.4%, and the accuracy rates to recognize edge and node are 94.1% and 67.9%, respectively. On average, it takes about 60 milliseconds to synthesize a program.

Keywords: program synthesis, flow chart, specification, graph recognition, CNN

Procedia PDF Downloads 119

Authors: Hamza Zidoum

Abstract:

This paper investigates the impact of human genetic variation on the function of human proteins using machine-learning algorithms. Single-Nucleotide Polymorphism represents the most common form of human genome variation. We focus on the single amino-acid polymorphism located in the coding region as they can affect the protein function leading to pathologic phenotypic change. We use several supervised Machine Learning methods to identify structural properties correlated with increased risk of the missense mutation being damaging. SVM associated with Principal Component Analysis give the best performance.

Keywords: single-nucleotide polymorphism, machine learning, feature selection, SVM

Procedia PDF Downloads 378

Authors: Netalie Shloim, Brian Brown, Siobhan Hugh-Jones, Jane Plastow, Diana Setiyawati, Anna Madill

Abstract:

In June 2021, the World Health Organization launched its guidance and technical packages on community mental health services, stressing a human rights-based approach to care. This initiative stems from an increasing acknowledgment of the role mental health plays in achieving the Sustainable Development Goals. Nevertheless, mental health remains a relatively neglected research area and the estimates for untreated mental disorders in low-and-middle-income countries (LMICs) are as high as 78% for adults. Moreover, the development sector and research programs too often side-line mental health as a privilege in the face of often immediate threats to life and livelihood. As a way of addressing this problem, this study aimed to examine past or ongoing GCRF projects to see if there were opportunities where mental health impact could have been achieved without compromising a study's main aim and without overburdening a project. Projects funded by the UKRI Global Challenges Research Fund (GCRF) were analyzed. This program was initiated in 2015 to support cutting-edge research that addresses the challenges faced by developing countries. By the end of May 2020, a total of 15,279 projects were funded of which only 3% had an explicit mental health focus. A sample of 36 non-mental-health-focused projects was then sampled for diversity across research council, challenge portfolio and world region. Each of these 36 projects was coded by two coders for opportunities to embed mental health impact. To facilitate coding, the literature was inspected for dimensions relevant to LMIC settings. Three main psychological and three main social dimensions were identified: promote a positive sense of self; promote positive emotions, safe expression and regulation of challenging emotions, coping strategies, and help-seeking; facilitate skills development; and facilitate community-building; preserve sociocultural identity; support community mobilization. Coding agreement was strong on missed opportunities for mental health impact on the three social dimensions: support community mobilization (92%), facilitate community building (83%), preserve socio-cultural identity (70%). Coding agreement was reasonably strong on missed opportunities for mental health impact on the three psychological dimensions: promote positive emotions (67%), facilitate skills development (61%), positive sense of self (58%). In order of frequency, the agreed perceived opportunities from the highest to lowest are: support community mobilization, facilitate community building, facilitate skills development, promote a positive sense of self, promote positive emotions, preserve sociocultural identity. All projects were considered to have an opportunity to support community mobilization and to facilitate skills development by at least one coder. Findings provided support that there were opportunities to embed mental health impact in research across the range of development sectors and identifies what kind of missed opportunities are most frequent. Hence, mainstreaming mental health has huge potential to tackle the lack of priority and funding it has attracted traditionally. The next steps are to understand the barriers to mainstreaming mental health and to work together to overcome them.

Keywords: GCRF, mental health, psychosocial wellbeing, LMIC

Procedia PDF Downloads 174

Authors: Deeksha Karunakar

Abstract:

The gaming industry is one of the biggest industries in the world. Video games are interactive works of authorship that require the execution of a computer programme on specialized hardware but which also incorporate a wide variety of other artistic mediums, such as music, scripts, stories, video, paintings, and characters, into which the player takes an active role. Therefore, video games are not made as singular, simple works but rather as a collection of elements that, if they reach a certain level of originality and creativity, can each be copyrighted on their own. A video game is made up of a wide variety of parts, all of which combine to form the overall sensation that we, the players, have while playing. The entirety of the components is implemented in the form of software code, which is then translated into the game's user interface. Even while copyright protection is already in place for the coding of software, the work that is produced because of that coding can also be protected by copyright. This includes the game's storyline or narrative, its characters, and even elements of the code on their own. In each sector, there is a potential legal framework required, and the gaming industry also requires legal frameworks. This represents the importance of intellectual property laws in each sector. This paper will explore the beginnings of video games, the various aspects of game copyrights, and the approach of the courts, including examples of a few different instances. Although the creative arts have always been known to draw inspiration from and build upon the works of others, it has not always been simple to evaluate whether a game has been cloned. The video game business is experiencing growth as it has never seen before today. The majority of today's video games are both pieces of software and works of audio-visual art. Even though the existing legal framework does not have a clause specifically addressing video games, it is clear that there is a great many alternative means by which this protection can be granted. This paper will represent the importance of copyright and trademark laws in the gaming industry and its regulations with the help of relevant case laws via utilizing doctrinal methodology to support its findings. The aim of the paper is to make aware of the applicability of intellectual property laws in the gaming industry and how the justice system is evolving to adapt to such new industries. Furthermore, it will provide in-depth knowledge of their relationship with each other.

Keywords: copyright, DMCA, gaming industry, trademark, WIPO

Procedia PDF Downloads 69

Authors: Rashedul Islam

Abstract:

A firmly-validated aim of teaching science is to support student enthusiasm for science learning with an outspread interest in scientific issues in future life. This is in keeping with the recent development in Gifted and Talented Education statement which instructs that gifted students have a renewed interest and natural aptitude in science. Yet, the practice of science teaching leaves many students with the feeling that science is difficult and compared to other school subjects, students interest in science is declining at the final years of the primary school. As a curriculum guides the teaching-learning activities in school, where significant consequences may result from the context of the curricula and syllabi, are a major feature of certain educational jurisdictions in NSW, Australia. The purpose of this study was an exploration of the curriculum sets the context to identify how science education is practiced through primary schools in Sydney, Australia. This phenomenon was explored through document review from two publicly available documents namely: the NSW Science Syllabus K-6, and Australian Curriculum: Foundation - 10 Science. To analyse the data, this qualitative study applied themed content analysis at three different levels, i.e., first cycle coding, second cycle coding- pattern codes, and thematic analysis. Preliminary analysis revealed the phenomenon of teaching-learning practices drawn from eight themes under three phenomena aligned with teachers’ practices and gifted student’s learning characteristics based on Gagné’s Differentiated Model of Gifted and Talent (DMGT). From the results, it appears that, overall, the two documents are relatively well-placed in terms of identifying the context of teaching and learning primary science to gifted students. However, educators need to make themselves aware of the ways in which the curriculum needs to be adapted to meet gifted students learning needs in science. It explores the important phenomena of teaching-learning context to provide gifted students with optimal educational practices including inquiry-based learning, problem-solving, open-ended tasks, creativity in science, higher order thinking, integration, and challenges. The significance of such a study lies in its potential to schools and further research in the field of gifted education.

Keywords: teaching primary science, gifted student learning, curriculum context, science syllabi, Australia

Procedia PDF Downloads 421

Authors: Farahnaz Sadat Golestan Hashemi, Mohd Razi Ismail, Mohd Y. Rafii

Abstract:

A huge revolution has emerged in genome engineering by the discovery of CRISPR (clustered regularly interspaced palindromic repeats) and CRISPR-associated system genes (Cas) in bacteria. The function of type II Streptococcus pyogenes (Sp) CRISPR/Cas9 system has been confirmed in various species. Other S. thermophilus (St) CRISPR-Cas systems, CRISPR1-Cas and CRISPR3-Cas, have been also reported for preventing phage infection. The CRISPR1-Cas system interferes by cleaving foreign dsDNA entering the cell in a length-specific and orientation-dependant manner. The S. thermophilus CRISPR3-Cas system also acts by cleaving phage dsDNA genomes at the same specific position inside the targeted protospacer as observed in the CRISPR1-Cas system. It is worth mentioning, for the effective DNA cleavage activity, RNA-guided Cas9 orthologs require their own specific PAM (protospacer adjacent motif) sequences. Activity levels are based on the sequence of the protospacer and specific combinations of favorable PAM bases. Therefore, based on the specific length and sequence of PAM followed by a constant length of target site for the three orthogonals of Cas9 protein, a well-organized procedure will be required for high-throughput and accurate mining of possible target sites in a large genomic dataset. Consequently, we created a reliable procedure to explore potential gRNA sequences for type I (Streptococcus thermophiles), II (Streptococcus pyogenes), and III (Streptococcus thermophiles) CRISPR/Cas systems. To mine CRISPR target sites, four different searching modes of sgRNA binding to target DNA strand were applied. These searching modes are as follows: i) coding strand searching, ii) anti-coding strand searching, iii) both strand searching, and iv) paired-gRNA searching. The output of such procedure highlights the power of comparative genome mining for different CRISPR/Cas systems. This could yield a repertoire of Cas9 variants with expanded capabilities of gRNA design, and will pave the way for further advance genome and epigenome engineering.

Keywords: CRISPR/Cas systems, gRNA mining, Streptococcus pyogenes, Streptococcus thermophiles

Procedia PDF Downloads 257

Authors: Prachi Desai, Ankita Gandhi, Mitali Acharya

Abstract:

Rainfall forecast is mainly used for predictions of rainfall in a specified area and determining their future rainfall conditions. Rainfall is always a global issue as it affects all major aspects of one's life. Agricultural, fisheries, forestry, tourism industry and other industries are widely affected by these conditions. The studies have resulted in insufficient availability of water resources and an increase in water demand in the near future. We already have a new forecast system that uses the deep Convolutional Neural Network (CNN) to forecast monthly rainfall and climate changes. We have also compared CNN against Artificial Neural Networks (ANN). Machine Learning techniques that are used in rainfall predictions include ARIMA Model, ANN, LR, SVM etc. The dataset on which we are experimenting is gathered online over the year 1901 to 20118. Test results have suggested more realistic improvements than conventional rainfall forecasts.

Keywords: ANN, CNN, supervised learning, machine learning, deep learning

Procedia PDF Downloads 202

Authors: Tsega Asresa, Getahun Tigistu, Melaku Bayih

Abstract:

Computer vision is a subfield of artificial intelligence that allows computers and systems to extract meaning from digital images and video. It is used in a wide range of fields of study, including self-driving cars, video surveillance, medical diagnosis, manufacturing, law, agriculture, quality control, health care, facial recognition, and military applications. Aromatic medicinal plants are botanical raw materials used in cosmetics, medicines, health foods, essential oils, decoration, cleaning, and other natural health products for therapeutic and Aromatic culinary purposes. These plants and their products not only serve as a valuable source of income for farmers and entrepreneurs but also going to export for valuable foreign currency exchange. In Ethiopia, there is a lack of technologies for the classification and identification of Aromatic medicinal plant parts and disease type cured by aromatic medicinal plants. Farmers, industry personnel, academicians, and pharmacists find it difficult to identify plant parts and disease types cured by plants before ingredient extraction in the laboratory. Manual plant identification is a time-consuming, labor-intensive, and lengthy process. To alleviate these challenges, few studies have been conducted in the area to address these issues. One way to overcome these problems is to develop a deep learning model for efficient identification of Aromatic medicinal plant parts with their corresponding disease type. The objective of the proposed study is to identify the aromatic medicinal plant parts and their disease type classification using computer vision technology. Therefore, this research initiated a model for the classification of aromatic medicinal plant parts and their disease type by exploring computer vision technology. Morphological characteristics are still the most important tools for the identification of plants. Leaves are the most widely used parts of plants besides roots, flowers, fruits, and latex. For this study, the researcher used RGB leaf images with a size of 128x128 x3. In this study, the researchers trained five cutting-edge models: convolutional neural network, Inception V3, Residual Neural Network, Mobile Network, and Visual Geometry Group. Those models were chosen after a comprehensive review of the best-performing models. The 80/20 percentage split is used to evaluate the model, and classification metrics are used to compare models. The pre-trained Inception V3 model outperforms well, with training and validation accuracy of 99.8% and 98.7%, respectively.

Keywords: aromatic medicinal plant, computer vision, convolutional neural network, deep learning, plant classification, residual neural network

Procedia PDF Downloads 187

Authors: Anup Anand Deshmukh, Catherine Soladie, Renaud Seguier

Abstract:

Emotion plays a key role in many applications like healthcare, to gather patients’ emotional behavior. Unlike typical ASR (Automated Speech Recognition) problems which focus on 'what was said', it is equally important to understand 'how it was said.' There are certain emotions which are given more importance due to their effectiveness in understanding human feelings. In this paper, we propose an approach that models human stress from audio signals. The research challenge in speech emotion detection is finding the appropriate set of acoustic features corresponding to an emotion. Another difficulty lies in defining the very meaning of emotion and being able to categorize it in a precise manner. Supervised Machine Learning models, including state of the art Deep Learning classification methods, rely on the availability of clean and labelled data. One of the problems in affective computation is the limited amount of annotated data. The existing labelled emotions datasets are highly subjective to the perception of the annotator. We address the first issue of feature selection by exploiting the use of traditional MFCC (Mel-Frequency Cepstral Coefficients) features in Convolutional Neural Network. Our proposed Emo-CNN (Emotion-CNN) architecture treats speech representations in a manner similar to how CNN’s treat images in a vision problem. Our experiments show that Emo-CNN consistently and significantly outperforms the popular existing methods over multiple datasets. It achieves 90.2% categorical accuracy on the Emo-DB dataset. We claim that Emo-CNN is robust to speaker variations and environmental distortions. The proposed approach achieves 85.5% speaker-dependant categorical accuracy for SAVEE (Surrey Audio-Visual Expressed Emotion) dataset, beating the existing CNN based approach by 10.2%. To tackle the second problem of subjectivity in stress labels, we use Lovheim’s cube, which is a 3-dimensional projection of emotions. Monoamine neurotransmitters are a type of chemical messengers in the brain that transmits signals on perceiving emotions. The cube aims at explaining the relationship between these neurotransmitters and the positions of emotions in 3D space. The learnt emotion representations from the Emo-CNN are mapped to the cube using three component PCA (Principal Component Analysis) which is then used to model human stress. This proposed approach not only circumvents the need for labelled stress data but also complies with the psychological theory of emotions given by Lovheim’s cube. We believe that this work is the first step towards creating a connection between Artificial Intelligence and the chemistry of human emotions.

Keywords: deep learning, brain chemistry, emotion perception, Lovheim's cube

Procedia PDF Downloads 154