Search results for: feature pre-processing

Authors: A: Annis Fathima, V. Vaidehi, S. Ajitha

Abstract:

Face recognition system finds many applications in surveillance and human computer interaction systems. As the applications using face recognition systems are of much importance and demand more accuracy, more robustness in the face recognition system is expected with less computation time. In this paper, a hybrid approach for face recognition combining Gabor Wavelet and Linear Discriminant Analysis (HGWLDA) is proposed. The normalized input grayscale image is approximated and reduced in dimension to lower the processing overhead for Gabor filters. This image is convolved with bank of Gabor filters with varying scales and orientations. LDA, a subspace analysis techniques are used to reduce the intra-class space and maximize the inter-class space. The techniques used are 2-dimensional Linear Discriminant Analysis (2D-LDA), 2-dimensional bidirectional LDA ((2D)2LDA), Weighted 2-dimensional bidirectional Linear Discriminant Analysis (Wt (2D)2 LDA). LDA reduces the feature dimension by extracting the features with greater variance. k-Nearest Neighbour (k-NN) classifier is used to classify and recognize the test image by comparing its feature with each of the training set features. The HGWLDA approach is robust against illumination conditions as the Gabor features are illumination invariant. This approach also aims at a better recognition rate using less number of features for varying expressions. The performance of the proposed HGWLDA approaches is evaluated using AT&T database, MIT-India face database and faces94 database. It is found that the proposed HGWLDA approach provides better results than the existing Gabor approach.

Keywords: face recognition, Gabor wavelet, LDA, k-NN classifier

Procedia PDF Downloads 450

Authors: Yuan-Hsiang Chang, Zhong-Xian Peng, Li-Der Jeng

Abstract:

Music has always been an integral part of human’s daily lives. But, for the most people, reading musical score and turning it into melody is not easy. This study aims to develop an Automatic music score recognition system using digital image processing, which can be used to read and analyze musical score images automatically. The technical approaches included: (1) staff region segmentation; (2) image preprocessing; (3) note recognition; and (4) accidental and rest recognition. Digital image processing techniques (e.g., horizontal /vertical projections, connected component labeling, morphological processing, template matching, etc.) were applied according to musical notes, accidents, and rests in staff notations. Preliminary results showed that our system could achieve detection and recognition rates of 96.3% and 91.7%, respectively. In conclusion, we presented an effective automated musical score recognition system that could be integrated in a system with a media player to play music/songs given input images of musical score. Ultimately, this system could also be incorporated in applications for mobile devices as a learning tool, such that a music player could learn to play music/songs.

Keywords: connected component labeling, image processing, morphological processing, optical musical recognition

Procedia PDF Downloads 394

Authors: Wei Yongxiang, Zhu Haoyue, Tang Heng, Yuan Qunwei

Abstract:

In order to study the preparation method of gear fatigue load spectrum for loaders, the load signal of four typical working conditions of loader is collected. The signal that reflects the law of load change is obtained by preprocessing the original signal. The torque of the drive axle is calculated by using the rain flow counting method. According to the operating time ratio of each working condition, the two-dimensional load spectrum based on the real working conditions of the drive axle of loader is established by the cycle extrapolation and synthesis method. The two-dimensional load spectrum is converted into one-dimensional load spectrum by means of the mean of torque equal damage method. Torque amplification includes the maximum load torque of the main reduction gear. Based on the theory of equal damage, the accelerated cycles are calculated. In this way, the load spectrum of the loading condition of the drive axle is prepared to reflect loading condition of the loader. The load spectrum can provide reference for fatigue life test and life prediction of loader drive axle.

Keywords: load spectrum, axle, torque, rain-flow counting method, extrapolation

Procedia PDF Downloads 342

Authors: Xiaocong Liu, Huazhen Wang, Ting He, Xiaozheng Li, Weihan Zhang, Jian Chen

Abstract:

The utilization of electronic medical record (EMR) data to establish the disease diagnosis model has become an important research content of biomedical informatics. Deep learning can automatically extract features from the massive data, which brings about breakthroughs in the study of EMR data. The challenge is that deep learning lacks semantic knowledge, which leads to impracticability in medical science. This research proposes a method of incorporating lexical-semantic knowledge from abundant entities into a convolutional neural network (CNN) framework for pediatric disease diagnosis. Firstly, medical terms are vectorized into Lexical Semantic Vectors (LSV), which are concatenated with the embedded word vectors of word2vec to enrich the feature representation. Secondly, the semantic distribution of medical terms serves as Semantic Decision Guide (SDG) for the optimization of deep learning models. The study evaluate the performance of LSV-SDG-CNN model on four kinds of Chinese EMR datasets. Additionally, CNN, LSV-CNN, and SDG-CNN are designed as baseline models for comparison. The experimental results show that LSV-SDG-CNN model outperforms baseline models on four kinds of Chinese EMR datasets. The best configuration of the model yielded an F1 score of 86.20%. The results clearly demonstrate that CNN has been effectively guided and optimized by lexical-semantic knowledge, and LSV-SDG-CNN model improves the disease classification accuracy with a clear margin.

Keywords: convolutional neural network, electronic medical record, feature representation, lexical semantics, semantic decision

Procedia PDF Downloads 112

Authors: Philipp Ruf, Massiwa Chabbi, Christoph Reich, Djaffar Ould-Abdeslam

Abstract:

In recent years, convolutional neural networks (CNN) have demonstrated high performance in image analysis, but oftentimes, there is only structured data available regarding a specific problem. By interpreting structured data as images, CNNs can effectively learn and extract valuable insights from tabular data, leading to improved predictive accuracy and uncovering hidden patterns that may not be apparent in traditional structured data analysis. In applying a single neural network for analyzing multimodal data, e.g., both structured and unstructured information, significant advantages in terms of time complexity and energy efficiency can be achieved. Converting structured data into images and merging them with existing visual material offers a promising solution for applying CNN in multimodal datasets, as they often occur in a medical context. By employing suitable preprocessing techniques, structured data is transformed into image representations, where the respective features are expressed as different formations of colors and shapes. In an additional step, these representations are fused with existing images to incorporate both types of information. This final image is finally analyzed using a CNN.

Keywords: CNN, image processing, tabular data, mixed dataset, data transformation, multimodal fusion

Procedia PDF Downloads 95

Authors: Hu Ding, Kai Liu, Guoan Tang

Abstract:

The Chinese Loess Plateau suffers from serious gully erosion induced by natural and human causes. Gully features detection including gully-affected area and its two dimension parameters (length, width, area et al.), is a significant task not only for researchers but also for policy-makers. This study aims at gully-affected area detection in three catchments of Chinese Loess Plateau, which were selected in Changwu, Ansai, and Suide by using unmanned aerial vehicle (UAV). The methodology includes a sequence of UAV data generation, image segmentation, feature calculation and selection, and random forest classification. Two experiments were conducted to investigate the influences of segmentation strategy and feature selection. Results showed that vertical and horizontal root-mean-square errors were below 0.5 and 0.2 m, respectively, which were ideal for the Loess Plateau region. The segmentation strategy adopted in this paper, which considers the topographic information, and optimal parameter combination can improve the segmentation results. Besides, the overall extraction accuracy in Changwu, Ansai, and Suide achieved was 84.62%, 86.46%, and 93.06%, respectively, which indicated that the proposed method for detecting gully-affected area is more objective and effective than traditional methods. This study demonstrated that UAV can bridge the gap between field measurement and satellite-based remote sensing, obtaining a balance in resolution and efficiency for catchment-scale gully erosion research.

Keywords: unmanned aerial vehicle (UAV), object-analysis image analysis, gully erosion, gully-affected area, Loess Plateau, random forest

Procedia PDF Downloads 196

Authors: Mona Hejazi, Ali Motie Nasrabadi

Abstract:

Epilepsy is a persistent neurological disorder that affects more than 50 million people worldwide. Hence, there is a necessity to introduce an efficient prediction model for making a correct diagnosis of the epileptic seizure and accurate prediction of its type. In this study we consider how the Effective Connectivity (EC) patterns obtained from intracranial Electroencephalographic (EEG) recordings reveal information about the dynamics of the epileptic brain and can be used to predict imminent seizures, as this will enable the patients (and caregivers) to take appropriate precautions. We use this definition because we believe that effective connectivity near seizures begin to change, so we can predict seizures according to this feature. Results are reported on the standard Freiburg EEG dataset which contains data from 21 patients suffering from medically intractable focal epilepsy. Six channels of EEG from each patients are considered and effective connectivity using Directed Transfer Function (DTF) and Granger Causality (GC) methods is estimated. We concentrate on effective connectivity standard deviation over time and feature changes in five brain frequency sub-bands (Alpha, Beta, Theta, Delta, and Gamma) are compared. The performance obtained for the proposed scheme in predicting seizures is: average prediction time is 50 minutes before seizure onset, the maximum sensitivity is approximate ~80% and the false positive rate is 0.33 FP/h. DTF method is more acceptable to predict epileptic seizures and generally we can observe that the greater results are in gamma and beta sub-bands. The research of this paper is significantly helpful for clinical applications, especially for the exploitation of online portable devices.

Keywords: effective connectivity, Granger causality, directed transfer function, epilepsy seizure prediction, EEG

Procedia PDF Downloads 442

Authors: Hongjian Jia

Abstract:

A finite-length cylindrical shell with a spherical cap is a typical engineering approximation model of actual underwater targets. The research on the omni-directional acoustic scattering characteristics of this target model can provide a favorable basis for the detection and identification of actual underwater targets. The elastic resonance characteristics of the target are the results of the comprehensive effect of the target length, shell-thickness ratio and materials. Under the conditions of different materials and geometric dimensions, the coincidence resonance characteristics of the target have obvious differences. Aiming at this problem, this paper obtains the omni-directional acoustic scattering field of the underwater hemispherical cylindrical shell by numerical calculation and studies the influence of target geometric parameters (length, shell-thickness ratio) and material parameters on the coincidence resonance characteristics of the target in turn. The study found that the formant interval is not a stable value and changes with the incident angle. Among them, the formant interval is less affected by the target length and shell-thickness ratio and is significantly affected by the material properties, which is an effective feature for classifying and identifying targets of different materials. The quadratic polynomial is utilized to fully fit the change relationship between the formant interval and the angle. The results show that the three fitting coefficients of the stainless steel and aluminum targets are significantly different, which can be used as an effective feature parameter to characterize the target materials.

Keywords: hemispherical cylindrical shell;, fine echo characteristics;, geometric and material parameters;, formant interval

Procedia PDF Downloads 83

Authors: Iman A. Saad, Loay E. George

Abstract:

Iris pattern is an important biological feature of human body; it becomes very hot topic in both research and practical applications. In this paper, an algorithm is proposed for iris recognition and a simple, efficient and fast method is introduced to extract a set of discriminatory features using first order gradient operator applied on grayscale images. The gradient based features are robust, up to certain extents, against the variations may occur in contrast or brightness of iris image samples; the variations are mostly occur due lightening differences and camera changes. At first, the iris region is located, after that it is remapped to a rectangular area of size 360x60 pixels. Also, a new method is proposed for detecting eyelash and eyelid points; it depends on making image statistical analysis, to mark the eyelash and eyelid as a noise points. In order to cover the features localization (variation), the rectangular iris image is partitioned into N overlapped sub-images (blocks); then from each block a set of different average directional gradient densities values is calculated to be used as texture features vector. The applied gradient operators are taken along the horizontal, vertical and diagonal directions. The low order norms of gradient components were used to establish the feature vector. Euclidean distance based classifier was used as a matching metric for determining the degree of similarity between the features vector extracted from the tested iris image and template features vectors stored in the database. Experimental tests were performed using 2639 iris images from CASIA V4-Interival database, the attained recognition accuracy has reached up to 99.92%.

Keywords: iris recognition, contrast stretching, gradient features, texture features, Euclidean metric

Procedia PDF Downloads 311

Authors: Wajih Abbassi, Zouhaier Ben Khelifa

Abstract:

The present research points towards the empirical validation of three options valuation models, the ad-hoc Black-Scholes model as proposed by Berkowitz (2001), the constant elasticity of variance model of Cox and Ross (1976) and the Kou jump-diffusion model (2002). Our empirical analysis has been conducted on a sample of 26,974 options written on three indexes, the S&P 500, Nasdaq 100 and the Russell 2000 that were negotiated during the year 2007 just before the sub-prime crisis. We start by presenting the theoretical foundations of the models of interest. Then we use the technique of trust-region-reflective algorithm to estimate the structural parameters of these models from cross-section of option prices. The empirical analysis shows the superiority of the Kou jump-diffusion model. This superiority arises from the ability of this model to portray the behavior of market participants and to be closest to the true distribution that characterizes the evolution of these indices. Indeed the double-exponential distribution covers three interesting properties that are: the leptokurtic feature, the memory less property and the psychological aspect of market participants. Numerous empirical studies have shown that markets tend to have both overreaction and under reaction over good and bad news respectively. Despite of these advantages there are not many empirical studies based on this model partly because probability distribution and option valuation formula are rather complicated. This paper is the first to have used the technique of nonlinear curve-fitting through the trust-region-reflective algorithm and cross-section options to estimate the structural parameters of the Kou jump-diffusion model.

Keywords: jump-diffusion process, Kou model, Leptokurtic feature, trust-region-reflective algorithm, US index options

Procedia PDF Downloads 407

Authors: Yulan Wu

Abstract:

The spread of fake news on social media has posed significant societal harm to the public and the nation, with its threats spanning various domains, including politics, economics, health, and more. News on social media often covers multiple domains, and existing models studied by researchers and relevant organizations often perform well on datasets from a single domain. However, when these methods are applied to social platforms with news spanning multiple domains, their performance significantly deteriorates. Existing research has attempted to enhance the detection performance of multi-domain datasets by adding single-domain labels to the data. However, these methods overlook the fact that a news article typically belongs to multiple domains, leading to the loss of domain knowledge information contained within the news text. To address this issue, research has found that news records in different domains often use different vocabularies to describe their content. In this paper, we propose a fake news detection framework that combines domain knowledge and expert knowledge. Firstly, it utilizes an unsupervised domain discovery module to generate a low-dimensional vector for each news article, representing domain embeddings, which can retain multi-domain knowledge of the news content. Then, a feature extraction module uses the domain embeddings discovered through unsupervised domain knowledge to guide multiple experts in extracting news knowledge for the total feature representation. Finally, a classifier is used to determine whether the news is fake or not. Experiments show that this approach can improve multi-domain fake news detection performance while reducing the cost of manually labeling domain labels.

Keywords: fake news, deep learning, natural language processing, multiple domains

Procedia PDF Downloads 47

Authors: Rita Magdalena, N. K. Caecar Pratiwi, Yunendah Nur Fuadah, Sofia Saidah, Bima Sakti

Abstract:

Vessel segmentation of retinal fundus is important for biomedical sciences in diagnosing ailments related to the eye. Segmentation can simplify medical experts in diagnosing retinal fundus image state. Therefore, in this study, we designed a software using MATLAB which enables the segmentation of the retinal blood vessels on retinal fundus images. There are two main steps in the process of segmentation. The first step is image preprocessing that aims to improve the quality of the image to be optimum segmented. The second step is the image segmentation in order to perform the extraction process to retrieve the retina’s blood vessel from the eye fundus image. The image segmentation methods that will be analyzed in this study are Morphology Operation, Discrete Wavelet Transform and combination of both. The amount of data that used in this project is 40 for the retinal image and 40 for manually segmentation image. After doing some testing scenarios, the average accuracy for Morphology Operation method is 88.46 % while for Discrete Wavelet Transform is 89.28 %. By combining the two methods mentioned in later, the average accuracy was increased to 89.53 %. The result of this study is an image processing system that can segment the blood vessels in retinal fundus with high accuracy and low computation time.

Keywords: discrete wavelet transform, fundus retina, morphology operation, segmentation, vessel

Procedia PDF Downloads 174

Authors: Aditya Karade, Sharada Falane, Dhananjay Deshmukh, Vijaykumar Mantri

Abstract:

Brain tumors pose a significant challenge in healthcare due to their complex nature and impact on patient outcomes. The application of deep learning (DL) algorithms in medical imaging have shown promise in accurate and efficient brain tumour detection. This paper explores the performance of various pre-trained DL models ResNet50, Xception, InceptionV3, EfficientNetB0, DenseNet121, NASNetMobile, VGG19, VGG16, and MobileNet on a brain tumour dataset sourced from Figshare. The dataset consists of MRI scans categorizing different types of brain tumours, including meningioma, pituitary, glioma, and no tumour. The study involves a comprehensive evaluation of these models’ accuracy and effectiveness in classifying brain tumour images. Data preprocessing, augmentation, and finetuning techniques are employed to optimize model performance. Among the evaluated deep learning models for brain tumour detection, ResNet50 emerges as the top performer with an accuracy of 98.86%. Following closely is Xception, exhibiting a strong accuracy of 97.33%. These models showcase robust capabilities in accurately classifying brain tumour images. On the other end of the spectrum, VGG16 trails with the lowest accuracy at 89.02%.

Keywords: brain tumour, MRI image, detecting and classifying tumour, pre-trained models, transfer learning, image segmentation, data augmentation

Procedia PDF Downloads 49

Authors: Cecile Laval, Harriet Lowe

Abstract:

Processing Instruction is a psycholinguistic pedagogical approach drawing insights from the Input Processing Model which establishes the initial innate strategies used by second language learners to connect form and meaning of linguistic features. With the ever-growing use of technology in Second Language Acquisition research, the present study uses eye-tracking to measure the effectiveness of Processing Instruction in the acquisition of French and its effects on learner’s cognitive strategies. The experiment was designed using a TOBII Pro-TX300 eye-tracker to measure participants’ default strategies when processing French linguistic input and any cognitive changes after receiving Processing Instruction treatment. Participants were drawn from lower intermediate adult learners of French at the University of Greenwich and randomly assigned to two groups. The study used a pre-test/post-test methodology. The pre-tests (one per linguistic item) were administered via the eye-tracker to both groups one week prior to instructional treatment. One group received full Processing Instruction treatment (explicit information on the grammatical item and on the processing strategies, and structured input activities) on the primary target linguistic feature (French past tense imperfective aspect). The second group received Processing Instruction treatment except the explicit information on the processing strategies. Three immediate post-tests on the three grammatical structures under investigation (French past tense imperfective aspect, French Subjunctive used for the expression of doubt, and the French causative construction with Faire) were administered with the eye-tracker. The eye-tracking data showed the positive change in learners’ processing of the French target features after instruction with improvement in the interpretation of the three linguistic features under investigation. 100% of participants in both groups made a statistically significant improvement (p=0.001) in the interpretation of the primary target feature (French past tense imperfective aspect) after treatment. 62.5% of participants made an improvement in the secondary target item (French Subjunctive used for the expression of doubt) and 37.5% of participants made an improvement in the cumulative target feature (French causative construction with Faire). Statistically there was no significant difference between the pre-test and post-test scores in the cumulative target feature; however, the variance approximately tripled between the pre-test and the post-test (3.9 pre-test and 9.6 post-test). This suggests that the treatment does not affect participants homogenously and implies a role for individual differences in the transfer-of-training effect of Processing Instruction. The use of eye-tracking provides an opportunity for the study of unconscious processing decisions made during moment-by-moment comprehension. The visual data from the eye-tracking demonstrates changes in participants’ processing strategies. Gaze plots from pre- and post-tests display participants fixation points changing from focusing on content words to focusing on the verb ending. This change in processing strategies can be clearly seen in the interpretation of sentences in both primary and secondary target features. This paper will present the research methodology, design and results of the experimental study using eye-tracking to investigate the primary effects and transfer-of-training effects of Processing Instruction. It will then provide evidence of the cognitive benefits of Processing Instruction in Second Language Acquisition and offer suggestion in second language teaching of grammar.

Keywords: eye-tracking, language teaching, processing instruction, second language acquisition

Procedia PDF Downloads 262

Authors: Gong Zhilin, Jing Yang, Jian Yin

Abstract:

The purpose of this paper is to develop a novel system for credit card fraud detection based on sequential modeling of data using hybrid deep learning models. The projected model encapsulates five major phases are pre-processing, imbalance-data handling, feature extraction, optimal feature selection, and fraud detection with an ensemble classifier. The collected raw data (input) is pre-processed to enhance the quality of the data through alleviation of the missing data, noisy data as well as null values. The pre-processed data are class imbalanced in nature, and therefore they are handled effectively with the K-means clustering-based SMOTE model. From the balanced class data, the most relevant features like improved Principal Component Analysis (PCA), statistical features (mean, median, standard deviation) and higher-order statistical features (skewness and kurtosis). Among the extracted features, the most optimal features are selected with the Self-improved Arithmetic Optimization Algorithm (SI-AOA). This SI-AOA model is the conceptual improvement of the standard Arithmetic Optimization Algorithm. The deep learning models like Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and optimized Quantum Deep Neural Network (QDNN). The LSTM and CNN are trained with the extracted optimal features. The outcomes from LSTM and CNN will enter as input to optimized QDNN that provides the final detection outcome. Since the QDNN is the ultimate detector, its weight function is fine-tuned with the Self-improved Arithmetic Optimization Algorithm (SI-AOA).

Keywords: credit card, data mining, fraud detection, money transactions

Procedia PDF Downloads 110

Authors: Alessandro Preziosi, Antonio Prioletti, Luca Castangia

Abstract:

Deformable part models achieve high precision in pedestrian recognition, but all publicly available implementations are too slow for real-time applications. We implemented a deformable part model algorithm fast enough for real-time use by exploiting information about the camera position and orientation. This implementation is both faster and more precise than alternative DPM implementations. These results are obtained by computing convolutions in the frequency domain and using lookup tables to speed up feature computation. This approach is almost an order of magnitude faster than the reference DPM implementation, with no loss in precision. Knowing the position of the camera with respect to horizon it is also possible prune many hypotheses based on their size and location. The range of acceptable sizes and positions is set by looking at the statistical distribution of bounding boxes in labelled images. With this approach it is not needed to compute the entire feature pyramid: for example higher resolution features are only needed near the horizon. This results in an increase in mean average precision of 5% and an increase in speed by a factor of two. Furthermore, to reduce misdetections involving small pedestrians near the horizon, input images are supersampled near the horizon. Supersampling the image at 1.5 times the original scale, results in an increase in precision of about 4%. The implementation was tested against the public KITTI dataset, obtaining an 8% improvement in mean average precision over the best performing DPM-based method. By allowing for a small loss in precision computational time can be easily brought down to our target of 100ms per image, reaching a solution that is faster and still more precise than all publicly available DPM implementations.

Keywords: autonomous vehicles, deformable part model, dpm, pedestrian detection, real time

Procedia PDF Downloads 258

Authors: K. N. Dinesh Babu, P. K. Gargava

Abstract:

Energization of a transformer results in sudden flow of current which is an effect of core magnetization. This current will be dominated by the presence of second harmonic, which in turn is used to segregate fault and inrush current, thus guaranteeing proper operation of the relay. This additional security in the relay sometimes obstructs or delays differential protection in a specific scenario, when the 2^nd harmonic content was present during a genuine fault. This kind of scenario can result in isolation of the transformer by Buchholz and pressure release valve (PRV) protection, which is acted when fault creates more damage in transformer. Such delays involve a huge impact on the insulation failure, and chances of repairing or rectifying fault of problem at site become very dismal. Sometimes this delay can cause fire in the transformer, and this situation becomes havoc for a sub-station. Such occurrences have been observed in field also when differential relay operation was delayed by 10-15 ms by second harmonic blocking in some specific conditions. These incidences have led to the need for an alternative solution to eradicate such unwarranted delay in operation in future. Modern numerical relay, called as intelligent electronic device (IED), is embedded with advanced protection features which permit higher flexibility and better provisions for tuning of protection logic and settings. Such flexibility in transformer protection IEDs, enables incorporation of alternative methods such as dynamic switching of second harmonic feature for blocking the differential protection with additional security. The analysis and precautionary measures carried out in this case, have been simulated and discussed in this paper to ensure that similar solutions can be adopted to inhibit analogous issues in future.

Keywords: differential protection, intelligent electronic device (IED), 2nd harmonic inhibit, inrush inhibit

Procedia PDF Downloads 268

Authors: Yufeng Li, Yinan Ma, Jing Wu, Chengnian Long

Abstract:

Infrared object detection aims at separating small and dim target from clutter background and its capabilities extend beyond the limits of visible light, making it invaluable in a wide range of applications such as improving safety, security, efficiency, and functionality. However, existing methods are usually sensitive to the noise of the input infrared image, leading to a decrease in target detection accuracy and an increase in the false alarm rate in high-noise environments. To address this issue, an infrared small target detection algorithm called YOLO-IR is proposed in this paper to improve the robustness to high infrared noise. To address the problem that high noise significantly reduces the clarity and reliability of target features in infrared images, we design a soft-threshold coordinate attention mechanism to improve the model’s ability to extract target features and its robustness to noise. Since the noise may overwhelm the local details of the target, resulting in the loss of small target features during depth down-sampling, we propose a deep and shallow feature fusion neck to improve the detection accuracy. In addition, because the generalized Intersection over Union (IoU)-based loss functions may be sensitive to noise and lead to unstable training in high-noise environments, we introduce a Wasserstein-distance based loss function to improve the training of the model. The experimental results show that YOLO-IR achieves a 5.0% improvement in recall and a 6.6% improvement in F1-score over existing state-of-art model.

Keywords: infrared small target detection, high noise, robustness, soft-threshold coordinate attention, feature fusion

Procedia PDF Downloads 36

Authors: Adrian Forca, Simeon Cainday III

Abstract:

The Electronic Payment Recording with Payment History Retrieval Module is developed intendedly for the College of Science and Technology. This system software innovates the manual process of recording the payments done in the department through the development of electronic payment recording system software shifting from the slow and time-consuming procedure to quick yet reliable and accurate way of recording payments because it immediately generates receipts for every transaction. As an added feature to its software process, generation of recorded payment report is integrated eliminating the manual reporting to a more easy and consolidated report. As an added feature to the system, all recorded payments of the students can be retrieved immediately making the system transparent and reliable payment recording software. Viewing the whole process, the system software will shift from the manual process to an organized software technology because the information will be stored in a logically correct and normalized database. Further, the software will be developed using the modern programming language and implement strict programming methods to validate all users accessing the system, evaluate all data passed into the system and information retrieved to ensure data accuracy and reliability. In addition, the system will identify the user and limit its access privilege to establish boundaries of the specific access to information allowed for the store, modify, and update making the information secure against unauthorized data manipulation. As a result, the System software will eliminate the manual procedure and replace with an innovative modern information technology resulting to the improvement of the whole process of payment recording fast, secure, accurate and reliable software innovations.

Keywords: collection, information system, manual procedure, payment

Procedia PDF Downloads 146

Authors: Zhongmin Wang, Wudong Fan, Hengshan Zhang, Yimin Zhou

Abstract:

In data-driven prognostic methods, the prediction accuracy of the estimation for remaining useful life of bearings mainly depends on the performance of health indicators, which are usually fused some statistical features extracted from vibrating signals. However, the existing health indicators have the following two drawbacks: (1) The differnet ranges of the statistical features have the different contributions to construct the health indicators, the expert knowledge is required to extract the features. (2) When convolutional neural networks are utilized to tackle time-frequency features of signals, the time-series of signals are not considered. To overcome these drawbacks, in this study, the method combining convolutional neural network with gated recurrent unit is proposed to extract the time-frequency image features. The extracted features are utilized to construct health indicator and predict remaining useful life of bearings. First, original signals are converted into time-frequency images by using continuous wavelet transform so as to form the original feature sets. Second, with convolutional and pooling layers of convolutional neural networks, the most sensitive features of time-frequency images are selected from the original feature sets. Finally, these selected features are fed into the gated recurrent unit to construct the health indicator. The results state that the proposed method shows the enhance performance than the related studies which have used the same bearing dataset provided by PRONOSTIA.

Keywords: continuous wavelet transform, convolution neural net-work, gated recurrent unit, health indicators, remaining useful life

Procedia PDF Downloads 109

Authors: Ali J. Ghandour, Houssam A. Krayem, Abedelkarim A. Jezzini

Abstract:

High-resolution satellite images and remote sensing can provide global information in a fast way compared to traditional methods of data collection. Under such high resolution, a road is not a thin line anymore. Objects such as cars and trees are easily identifiable. Automatic vehicles enumeration can be considered one of the most important applications in traffic management. In this paper, autonomous vehicle detection and classification approach in highway environment is proposed. This approach consists mainly of three stages: (i) first, a set of preprocessing operations are applied including soil, vegetation, water suppression. (ii) Then, road networks detection and delineation is implemented using built-up area index, followed by several morphological operations. This step plays an important role in increasing the overall detection accuracy since vehicles candidates are objects contained within the road networks only. (iii) Multi-level Otsu segmentation is implemented in the last stage, resulting in vehicle detection and classification, where detected vehicles are classified into cars and trucks. Accuracy assessment analysis is conducted over different study areas to show the great efficiency of the proposed method, especially in highway environment.

Keywords: remote sensing, object identification, vehicle and road extraction, vehicle and road features-based classification

Procedia PDF Downloads 210

Authors: Nonlapun Meenil, Pisitpong Intarapong, Thitima Wongsheree, Pranchalee Samanpiboon

Abstract:

In Thailand, once pineapples are harvested, they must be classified into two classes based on their sweetness: sweet and unsweet. This paper has studied and developed the assessment of internal quality of pineapples using a low-cost compact spectroscopy sensor according to the Spectroscopy approach and Neural Network (NN). During the experiments, Batavia pineapples were utilized, generating 100 samples. The extracted pineapple juice of each sample was used to determine the Soluble Solid Content (SSC) labeling into sweet and unsweet classes. In terms of experimental equipment, the sensor cover was specifically designed to install the sensor and light source to read the reflectance at a five mm depth from pineapple flesh. By using a spectroscopy sensor, data on visible and near-infrared reflectance (Vis-NIR) were collected. The NN was used to classify the pineapple classes. Before the classification step, the preprocessing methods, which are Class balancing, Data shuffling, and Standardization were applied. The 510 nm and 900 nm reflectance values of the middle parts of pineapples were used as features of the NN. With the Sequential model and Relu activation function, 100% accuracy of the training set and 76.67% accuracy of the test set were achieved. According to the abovementioned information, using a low-cost compact spectroscopy sensor has achieved favorable results in classifying the sweetness of the two classes of pineapples.

Keywords: neural network, pineapple, soluble solid content, spectroscopy

Procedia PDF Downloads 50

Authors: Varun Agarwal

Abstract:

Introduction: With the advent of personalized medicine, histopathological review of whole slide images (WSIs) for cancer diagnosis presents an exceedingly time-consuming, complex task. Specifically, detecting metastatic regions in WSIs of sentinel lymph node biopsies necessitates a full-scanned, holistic evaluation of the image. Thus, digital pathology, low-level image manipulation algorithms, and machine learning provide significant advancements in improving the efficiency and accuracy of WSI analysis. Using Camelyon16 data, this paper proposes a deep learning pipeline to automate and ameliorate breast cancer metastasis localization and WSI classification. Methodology: The model broadly follows five stages -region of interest detection, WSI partitioning into image tiles, convolutional neural network (CNN) image-segment classifications, probabilistic mapping of tumor localizations, and further processing for whole WSI classification. Transfer learning is applied to the task, with the implementation of Inception-ResNetV2 - an effective CNN classifier that uses residual connections to enhance feature representation, adding convolved outputs in the inception unit to the proceeding input data. Moreover, in order to augment the performance of the transfer learning CNN, a stack of convolutional denoising autoencoders (CDAE) is applied to produce embeddings that enrich image representation. Through a saliency-detection algorithm, visual training segments are generated, which are then processed through a denoising autoencoder -primarily consisting of convolutional, leaky rectified linear unit, and batch normalization layers- and subsequently a contrast-normalization function. A spatial pyramid pooling algorithm extracts the key features from the processed image, creating a viable feature map for the CNN that minimizes spatial resolution and noise. Results and Conclusion: The simplified and effective architecture of the fine-tuned transfer learning Inception-ResNetV2 network enhanced with the CDAE stack yields state of the art performance in WSI classification and tumor localization, achieving AUC scores of 0.947 and 0.753, respectively. The convolutional feature retention and compilation with the residual connections to inception units synergized with the input denoising algorithm enable the pipeline to serve as an effective, efficient tool in the histopathological review of WSIs.

Keywords: breast cancer, convolutional neural networks, metastasis mapping, whole slide images

Procedia PDF Downloads 110

Authors: Dong Tran, Thanh Dac Van, Ly Le

Abstract:

Hemagglutinin (HA) and Neuraminidase (NA) play an important role in host immune evasion across influenza virus evolution process. The correlation between HA and NA evolution in respect to epitopic evolution and drug interaction has yet to be investigated. In this study, combining of sequence to structure evolution and statistical analysis on epitopic/binding site specificity, we identified potential therapeutic features of HA and NA that show specific antibody binding site of HA and specific binding distribution within NA active site of current inhibitors. Our approach introduces the use of sequence variation and molecular interaction to provide an effective strategy in establishing experimental based distributed representations of protein-protein/ligand complexes. The most important advantage of our method is that it does not require complete dataset of complexes but rather directly inferring feature interaction from sequence variation and molecular interaction. Using correlated sequence analysis, we additionally identified co-evolved mutations associated with maintaining HA/NA structural and functional variability toward immunity and therapeutic treatment. Our investigation on the HA binding specificity revealed unique conserved stalk domain interacts with unique loop domain of universal antibodies (CR9114, CT149, CR8043, CR8020, F16v3, CR6261, F10). On the other hand, NA inhibitors (Oseltamivir, Zaninamivir, Laninamivir) showed specific conserved residue contribution and similar to that of NA substrate (sialic acid) which can be exploited for drug design. Our study provides an important insight into rational design and identification of novel therapeutics targeting universally recognized feature of influenza HA/NA.

Keywords: influenza virus, hemagglutinin (HA), neuraminidase (NA), sequence evolution

Procedia PDF Downloads 138

Authors: Sujata Kulkarni, Udhav Bhosle, Vijaykumar T.

Abstract:

Due to the closeness between seismic signals and non-seismic signals, it is vital to detect earthquakes using conventional methods. In order to distinguish between seismic events and non-seismic events depending on their amplitude, our study processes the data that come from seismic sensors. The authors suggest a robust noise suppression technique that makes use of a bandpass filter, an IIR Wiener filter, recursive short-term average/long-term average (STA/LTA), and Carl short-term average (STA)/long-term average for event identification (LTA). The trigger ratio used in the proposed study to differentiate between seismic and non-seismic activity is determined. The proposed work focuses on significant feature extraction for machine learning-based seismic event detection. This serves as motivation for compiling a dataset of all features for the identification and forecasting of seismic signals. We place a focus on feature vector dimension reduction techniques due to the temporal complexity. The proposed notable features were experimentally tested using a machine learning model, and the results on unseen data are optimal. Finally, a presentation using a hybrid dataset (captured by different sensors) demonstrates how this model may also be employed in a real-time setting while lowering false alarm rates. The planned study is based on the examination of seismic signals obtained from both individual sensors and sensor networks (SN). A wideband seismic signal from BSVK and CUKG station sensors, respectively located near Basavakalyan, Karnataka, and the Central University of Karnataka, makes up the experimental dataset.

Keywords: Carl STA/LTA, features extraction, real time, dataset, machine learning, seismic detection

Procedia PDF Downloads 93

Authors: Takatoshi Ishii, Koji Kimita, Keiichi Muramatsu, Yoshiki Shimomura

Abstract:

To increase the quality of learning, teacher and learner need mutual effort for realization of educational value. For this purpose, we need to manage the co-creational education among teacher and learners. In this research, we try to find a feature of co-creational education. To be more precise, we analyzed learners’ reports by natural language processing, and extract some features that describe the state of the co-creational education.

Keywords: co-creational education, e-portfolios, ICT integration, latent dirichlet allocation

Procedia PDF Downloads 596

Authors: Abeer A. Aljohani

Abstract:

COVID-19 virus spread has been one of the extreme pandemics across the globe. It is also referred to as coronavirus, which is a contagious disease that continuously mutates into numerous variants. Currently, the B.1.1.529 variant labeled as omicron is detected in South Africa. The huge spread of COVID-19 disease has affected several lives and has surged exceptional pressure on the healthcare systems worldwide. Also, everyday life and the global economy have been at stake. This research aims to predict COVID-19 disease in its initial stage to reduce the death count. Machine learning (ML) is nowadays used in almost every area. Numerous COVID-19 cases have produced a huge burden on the hospitals as well as health workers. To reduce this burden, this paper predicts COVID-19 disease is based on the symptoms and medical history of the patient. This research presents a unique architecture for COVID-19 detection using ML techniques integrated with feature dimensionality reduction. This paper uses a standard UCI dataset for predicting COVID-19 disease. This dataset comprises symptoms of 5434 patients. This paper also compares several supervised ML techniques to the presented architecture. The architecture has also utilized 10-fold cross validation process for generalization and the principal component analysis (PCA) technique for feature reduction. Standard parameters are used to evaluate the proposed architecture including F1-Score, precision, accuracy, recall, receiver operating characteristic (ROC), and area under curve (AUC). The results depict that decision tree, random forest, and neural networks outperform all other state-of-the-art ML techniques. This achieved result can help effectively in identifying COVID-19 infection cases.

Keywords: supervised machine learning, COVID-19 prediction, healthcare analytics, random forest, neural network

Procedia PDF Downloads 69

Authors: Yuan-Hsiang Chang, Pin-Chi Lin, Li-Der Jeng

Abstract:

Advance in techniques of image and video processing has enabled the development of intelligent video surveillance systems. This study was aimed to automatically detect moving human objects and to analyze events of dual human interaction in a surveillance scene. Our system was developed in four major steps: image preprocessing, human object detection, human object tracking, and motion trajectory analysis. The adaptive background subtraction and image processing techniques were used to detect and track moving human objects. To solve the occlusion problem during the interaction, the Kalman filter was used to retain a complete trajectory for each human object. Finally, the motion trajectory analysis was developed to distinguish between the interaction and non-interaction events based on derivatives of trajectories related to the speed of the moving objects. Using a database of 60 video sequences, our system could achieve the classification accuracy of 80% in interaction events and 95% in non-interaction events, respectively. In summary, we have explored the idea to investigate a system for the automatic classification of events for interaction and non-interaction events using surveillance cameras. Ultimately, this system could be incorporated in an intelligent surveillance system for the detection and/or classification of abnormal or criminal events (e.g., theft, snatch, fighting, etc.).

Keywords: motion detection, motion tracking, trajectory analysis, video surveillance

Procedia PDF Downloads 519

Authors: Zamri Abu Bakar Zamri, Normaly Kamal Ismail Normaly, Mohd Izani Mohamed Rawi Izani

Abstract:

Malay noun compound are phrases that consist of two or more nouns. The key characteristic behind noun compounds lies on its frequent occurrences within the text. Therefore, extracting these noun compounds is essential for several domains of research such as Information Retrieval, Sentiment Analysis and Question Answering. Many research efforts have been proposed in terms of extracting Malay noun compounds using linguistic and statistical approaches. Most of the existing methods have concentrated on the extraction of bi-gram noun+noun compound. However, extracting noun+verb, noun+adjective and noun+prepositional is challenging due to the difficulty of selecting an appropriate method with effective results. Thus, there is still room for improvement in terms of enhancing the effectiveness of compound word extraction. Therefore, this study proposed a combination of linguistic approach and statistical measures in order to enhance the extraction of compound words. Several preprocessing steps are involved including normalization, tokenization, and stemming. The linguistic approach that has been used in this study is Part-of-Speech (POS) tagging. In addition, a new linguistic pattern for named entities has been utilized using a list of Malays named entities in order to enhance the linguistic approach in terms of noun compound recognition. The proposed statistical measures consists of NC-value, NTC-value and NLC value.

Keywords: Compound Word, Noun Compound, Linguistic Approach, Statistical Approach

Procedia PDF Downloads 325

Authors: Sang-Keun Moon, Hong-Rok Lim, Jin-O Kim

Abstract:

This paper presents a Power Quality (PQ) modeling and filtering processes for the distribution system disturbances using recognition learning methods. Typical PQ waveforms with mathematical applications and gathered field data are applied to the proposed models. The objective of this paper is analyzing PQ data with respect to monitoring, discriminating, and evaluating the waveform of power disturbances to ensure the system preventative system failure protections and complex system problem estimations. Examined signal filtering techniques are used for the field waveform noises and feature extractions. Using extraction and learning classification techniques, the efficiency was verified for the recognition of the PQ disturbances with focusing on interactive modeling methods in this paper. The waveform of selected 8 disturbances is modeled with randomized parameters of IEEE 1159 PQ ranges. The range, parameters, and weights are updated regarding field waveform obtained. Along with voltages, currents have same process to obtain the waveform features as the voltage apart from some of ratings and filters. Changing loads are causing the distortion in the voltage waveform due to the drawing of the different patterns of current variation. In the conclusion, PQ disturbances in the voltage and current waveforms indicate different types of patterns of variations and disturbance, and a modified technique based on the symmetrical components in time domain was proposed in this paper for the PQ disturbances detection and then classification. Our method is based on the fact that obtained waveforms from suggested trigger conditions contain potential information for abnormality detections. The extracted features are sequentially applied to estimation and recognition learning modules for further studies.

Keywords: power quality recognition, PQ modeling, waveform feature extraction, disturbance trigger condition, PQ signal filtering

Procedia PDF Downloads 166