Search results for: deep convolutional features

Authors: Shibo Wei, Ting Jiang

Abstract:

Convolutional-recurrent neural networks (CRN) have achieved much success recently in the speech enhancement field. The common processing method is to use the convolution layer to compress the feature space by multiple upsampling and then model the compressed features with the LSTM layer. At last, the enhanced speech is obtained by deconvolution operation to integrate the global information of the speech sequence. However, the feature space compression process may cause the loss of information, so we propose to model the upsampling result of each step with the residual LSTM layer, then join it with the output of the deconvolution layer and input them to the next deconvolution layer, by this way, we want to integrate the global information of speech sequence better. The experimental results show the network model (RES-CRN) we introduce can achieve better performance than LSTM without residual and overlaying LSTM simply in the original CRN in terms of scale-invariant signal-to-distortion ratio (SI-SNR), speech quality (PESQ), and intelligibility (STOI).

Keywords: convolutional-recurrent neural networks, speech enhancement, residual LSTM, SI-SNR

Procedia PDF Downloads 171

Authors: Parisa Mansour

Abstract:

Cystic fibrosis (CF) is one of the most common autosomal recessive diseases among whites. It mostly affects the lungs, causing infections and inflammation that account for 90% of deaths in CF patients. Because of this high variability in clinical presentation and organ involvement, investigating treatment responses and evaluating lung changes over time is critical to preventing CF progression. High-resolution computed tomography (HRCT) greatly facilitates the assessment of lung disease progression in CF patients. Recently, artificial intelligence was used to analyze chest CT scans of CF patients. In this paper, we propose a convolutional neural network (CNN) approach to classify CF lung patterns in HRCT images. The proposed network consists of two convolutional layers with 3 × 3 kernels and maximally connected in each layer, followed by two dense layers with 1024 and 10 neurons, respectively. The softmax layer prepares a predicted output probability distribution between classes. This layer has three exits corresponding to the categories of normal (healthy), bronchitis and inflammation. To train and evaluate the network, we constructed a patch-based dataset extracted from more than 1100 lung HRCT slices obtained from 45 CF patients. Comparative evaluation showed the effectiveness of the proposed CNN compared to its close peers. Classification accuracy, average sensitivity and specificity of 93.64%, 93.47% and 96.61% were achieved, indicating the potential of CNNs in analyzing lung CF patterns and monitoring lung health. In addition, the visual features extracted by our proposed method can be useful for automatic measurement and finally evaluation of the severity of CF patterns in lung HRCT images.

Keywords: HRCT, CF, cystic fibrosis, chest CT, artificial intelligence

Procedia PDF Downloads 36

Authors: Lanchi Xie, Zhihui Li, Zhigang Li, Guiqiang Wang, Lei Xu, Yuwen Yan

Abstract:

Image-based facial features can be classified into category recognition features and individual recognition features. Current automated face recognition systems extract a specific feature vector of different dimensions from a facial image according to their pre-trained neural network. However, to improve the efficiency of parameter calculation, an algorithm generally reduces the image details by pooling. The operation will overlook the details concerned much by forensic experts. In our experiment, we adopted a variety of face recognition algorithms based on deep learning, compared a large number of naturally collected face images with the known data of the same person's frontal ID photos. Downscaling and manual handling were performed on the testing images. The results supported that the facial recognition algorithms based on deep learning detected structural and morphological information and rarely focused on specific markers such as stains and moles. Overall performance, distribution of genuine scores and impostor scores, and likelihood ratios were tested to evaluate the accuracy of biometric systems and forensic experts. Experiments showed that the biometric systems were skilled in distinguishing category features, and forensic experts were better at discovering the individual features of human faces. In the proposed approach, a fusion was performed at the score level. At the specified false accept rate, the framework achieved a lower false reject rate. This paper contributes to improving the interpretability of the objective method of facial comparison and provides a novel method for human-machine collaboration in this field.

Keywords: likelihood ratio, automated facial recognition, facial comparison, biometrics

Procedia PDF Downloads 100

Authors: Pengfei Meng, Shuangcheng Jia, Qian Li

Abstract:

We proposed a real-time high-precision semantic segmentation network based on a multi-resolution feature fusion module, the auxiliary feature extracting module, upsampling module, and atrous spatial pyramid pooling (ASPP) module. We designed a feature fusion structure, which is integrated with sufficient features of different resolutions. We also studied the effect of side-branch structure on the network and made discoveries. Based on the discoveries about the side-branch of the network structure, we used a side-branch auxiliary feature extraction layer in the network to improve the effectiveness of the network. We also designed upsampling module, which has better results than the original upsampling module. In addition, we also re-considered the locations and number of atrous spatial pyramid pooling (ASPP) modules and modified the network structure according to the experimental results to further improve the effectiveness of the network. The network presented in this paper takes the backbone network of Bisenetv2 as a basic network, based on which we constructed a network structure on which we made improvements. We named this network deep multiple-resolution bilateral networks for real-time, referred to as DMBR-Net. After experimental testing, our proposed DMBR-Net network achieved 81.2% mIoU at 119FPS on the Cityscapes validation dataset, 80.7% mIoU at 109FPS on the CamVid test dataset, 29.9% mIoU at 78FPS on the COCOStuff test dataset. Compared with all lightweight real-time semantic segmentation networks, our network achieves the highest accuracy at an appropriate speed.

Keywords: multi-resolution feature fusion, atrous convolutional, bilateral networks, pyramid pooling

Procedia PDF Downloads 108

Authors: Qasem Qananwah, Ahmad Dagamseh, Hiam AlQuran, Khalid Shaker Ibrahim

Abstract:

Blood pressure helps the physicians greatly to have a deep insight into the cardiovascular system. The determination of individual blood pressure is a standard clinical procedure considered for cardiovascular system problems. The conventional techniques to measure blood pressure (e.g. cuff method) allows a limited number of readings for a certain period (e.g. every 5-10 minutes). Additionally, these systems cause turbulence to blood flow; impeding continuous blood pressure monitoring, especially in emergency cases or critically ill persons. In this paper, the most important statistical features in the photoplethysmogram (PPG) signals were extracted to estimate the blood pressure noninvasively. PPG signals from more than 40 subjects were measured and analyzed and 12 features were extracted. The features were fed to principal component analysis (PCA) to find the most important independent features that have the highest correlation with blood pressure. The results show that the stiffness index means and standard deviation for the beat-to-beat heart rate were the most important features. A model representing both features for Systolic Blood Pressure (SBP) and Diastolic Blood Pressure (DBP) was obtained using a statistical regression technique. Surface fitting is used to best fit the series of data and the results show that the error value in estimating the SBP is 4.95% and in estimating the DBP is 3.99%.

Keywords: blood pressure, noninvasive optical system, principal component analysis, PCA, continuous monitoring

Procedia PDF Downloads 132

Authors: P. Karthick, K. Mahesh

Abstract:

Video has become an increasingly significant component of our digital everyday contact. With the advancement of greater contents and shows of the resolution, its significant volume poses serious obstacles to the objective of receiving, distributing, compressing, and revealing video content of high quality. In this paper, we propose the primary beginning to complete a deep video compression model that jointly upgrades all video compression components. The video compression method involves splitting the video into frames, comparing the images using convolutional neural networks (CNN) to remove duplicates, repeating the single image instead of the duplicate images by recognizing and detecting minute changes using generative adversarial network (GAN) and recorded with long short-term memory (LSTM). Instead of the complete image, the small changes generated using GAN are substituted, which helps in frame level compression. Pixel wise comparison is performed using K-nearest neighbours (KNN) over the frame, clustered with K-means, and singular value decomposition (SVD) is applied for each and every frame in the video for all three color channels [Red, Green, Blue] to decrease the dimension of the utility matrix [R, G, B] by extracting its latent factors. Video frames are packed with parameters with the aid of a codec and converted to video format, and the results are compared with the original video. Repeated experiments on several videos with different sizes, duration, frames per second (FPS), and quality results demonstrate a significant resampling rate. On average, the result produced had approximately a 10% deviation in quality and more than 50% in size when compared with the original video.

Keywords: video compression, K-means clustering, convolutional neural network, generative adversarial network, singular value decomposition, pixel visualization, stochastic gradient descent, frame per second extraction, RGB channel extraction, self-detection and deciding system

Procedia PDF Downloads 161

Authors: Gurukiran Veerapur, Nytik Birudavolu, Seetharam U. N., Chandravva Hebbi, R. Praneeth Reddy

Abstract:

This work presents a robust approach for the recognition of characters in Telugu and Kannada, two South Indian scripts with structural similarities in characters. To recognize the characters exhaustive datasets are required, but there are only a few publicly available datasets. As a result, we decided to create a dataset for one language (source language),train the model with it, and then test it with the target language.Telugu is the target language in this work, whereas Kannada is the source language. The suggested method makes use of Canny edge features to increase character identification accuracy on pictures with noise and different lighting. A dataset of 45,150 images containing printed Kannada characters was created. The Nudi software was used to automatically generate printed Kannada characters with different writing styles and variations. Manual labelling was employed to ensure the accuracy of the character labels. The deep learning models like CNN (Convolutional Neural Network) and Visual Attention neural network (VAN) are used to experiment with the dataset. A Visual Attention neural network (VAN) architecture was adopted, incorporating additional channels for Canny edge features as the results obtained were good with this approach. The model's accuracy on the combined Telugu and Kannada test dataset was an outstanding 97.3%. Performance was better with Canny edge characteristics applied than with a model that solely used the original grayscale images. The accuracy of the model was found to be 80.11% for Telugu characters and 98.01% for Kannada words when it was tested with these languages. This model, which makes use of cutting-edge machine learning techniques, shows excellent accuracy when identifying and categorizing characters from these scripts.

Keywords: base characters, modifiers, guninthalu, aksharas, vattakshara, VAN

Procedia PDF Downloads 26

Authors: Muhammad Atif, Cang Yan

Abstract:

The enhancement of low-light images is a significant area of study aimed at enhancing the quality of captured images in challenging lighting environments. Recently, methods based on convolutional neural networks (CNN) have gained prominence as they offer state-of-the-art performance. However, many approaches based on CNN rely on increasing the size and complexity of the neural network. In this study, we propose an alternative method for improving low-light images using an autoencoder-based multiscale knowledge transfer model. Our method leverages the power of three autoencoders, where the encoders of the first two autoencoders are directly connected to the decoder of the third autoencoder. Additionally, the decoder of the first two autoencoders is connected to the encoder of the third autoencoder. This architecture enables effective knowledge transfer, allowing the third autoencoder to learn and benefit from the enhanced knowledge extracted by the first two autoencoders. We further integrate the proposed model into the PIX to PIX GAN framework. By integrating our proposed model as the generator in the GAN framework, we aim to produce enhanced images that not only exhibit improved visual quality but also possess a more authentic and realistic appearance. These experimental results, both qualitative and quantitative, show that our method is better than the state-of-the-art methodologies.

Keywords: low light image enhancement, deep learning, convolutional neural network, image processing

Procedia PDF Downloads 34

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 178

Authors: Mohamed Raouf Benmakrelouf, Wafa Karouche, Joseph Rynkiewicz

Abstract:

In this paper, we propose an alternative method to construct a Bayesian Network (BN). This method relies on a convolutional neural network (CNN classifier), which determinates the edges of the network skeleton. We train a CNN on a normalized empirical probability density distribution (NEPDF) for predicting causal interactions and relationships. We have to find the optimal Bayesian network structure for causal inference. Indeed, we are undertaking a search for pair-wise causality, depending on considered causal assumptions. In order to avoid unreasonable causal structure, we consider a blacklist and a whitelist of causality senses. We tested the method on real data to assess the influence of education on the voting intention for the extreme right-wing party. We show that, with this method, we get a safer causal structure of variables (Bayesian Network) and make to identify a variable that satisfies the backdoor criterion.

Keywords: Bayesian network, structure learning, optimal search, convolutional neural network, causal inference

Procedia PDF Downloads 142

Authors: A. Degale Desta, Cheng Jian

Abstract:

Deep learning application in computer vision is rapidly advancing, giving it the ability to monitor the public and quickly identify potentially anomalous behaviour from crowd scenes. Therefore, the purpose of the current work is to improve the performance of safety of people in crowd events from panic behaviour through introducing the innovative idea of Aggregation of Ensembles (AOE), which makes use of the pre-trained ConvNets and a pool of classifiers to find anomalies in video data with packed scenes. According to the theory of algorithms that applied K-means, KNN, CNN, SVD, and Faster-CNN, YOLOv5 architectures learn different levels of semantic representation from crowd videos; the proposed approach leverages an ensemble of various fine-tuned convolutional neural networks (CNN), allowing for the extraction of enriched feature sets. In addition to the above algorithms, a long short-term memory neural network to forecast future feature values and a handmade feature that takes into consideration the peculiarities of the crowd to understand human behavior. On well-known datasets of panic situations, experiments are run to assess the effectiveness and precision of the suggested method. Results reveal that, compared to state-of-the-art methodologies, the system produces better and more promising results in terms of accuracy and processing speed.

Keywords: action recognition, computer vision, crowd detecting and tracking, deep learning

Procedia PDF Downloads 123

Authors: Shreeyam, Ranjan Kumar Sah, Shivangi

Abstract:

Recognizing and solving handwritten mathematical expressions can be a challenging task, particularly when certain characters are segmented and classified. This project proposes a solution that uses Convolutional Neural Network (CNN) and image processing techniques to accurately solve various types of equations, including arithmetic, quadratic, and trigonometric equations, as well as logical operations like logical AND, OR, NOT, NAND, XOR, and NOR. The proposed solution also provides a graphical solution, allowing users to visualize equations and their solutions. In addition to equation solving, the platform, called CNNCalc, offers a comprehensive learning experience for students. It provides educational content, a quiz platform, and a coding platform for practicing programming skills in different languages like C, Python, and Java. This all-in-one solution makes the learning process engaging and enjoyable for students. The proposed methodology includes horizontal compact projection analysis and survey for segmentation and binarization, as well as connected component analysis and integrated connected component analysis for character classification. The compact projection algorithm compresses the horizontal projections to remove noise and obtain a clearer image, contributing to the accuracy of character segmentation. Experimental results demonstrate the effectiveness of the proposed solution in solving a wide range of mathematical equations. CNNCalc provides a powerful and user-friendly platform for solving equations, learning, and practicing programming skills. With its comprehensive features and accurate results, CNNCalc is poised to revolutionize the way students learn and solve mathematical equations. The platform utilizes a custom-designed Convolutional Neural Network (CNN) with image processing techniques to accurately recognize and classify symbols within handwritten equations. The compact projection algorithm effectively removes noise from horizontal projections, leading to clearer images and improved character segmentation. Experimental results demonstrate the accuracy and effectiveness of the proposed solution in solving a wide range of equations, including arithmetic, quadratic, trigonometric, and logical operations. CNNCalc features a user-friendly interface with a graphical representation of equations being solved, making it an interactive and engaging learning experience for users. The platform also includes tutorials, testing capabilities, and programming features in languages such as C, Python, and Java. Users can track their progress and work towards improving their skills. CNNCalc is poised to revolutionize the way students learn and solve mathematical equations with its comprehensive features and accurate results.

Keywords: AL, ML, hand written equation solver, maths, computer, CNNCalc, convolutional neural networks

Procedia PDF Downloads 82

Authors: Guantao Bai

Abstract:

Public space is an indispensable and important component of the urban built environment. How to more accurately evaluate the usage characteristics of public space can help improve its spatial quality. Compared to traditional survey methods, computer vision technology based on deep learning has advantages such as dynamic observation and low cost. This study takes the public space of Shaping Park as an example and, based on deep learning computer vision technology, processes and analyzes the image data of the public space to obtain the spatial usage characteristics and spatiotemporal characteristics of the public space. Research has found that the spontaneous activity time in public spaces is relatively random with a relatively short average activity time, while social activities have a relatively stable activity time with a longer average activity time. Computer vision technology based on deep learning can effectively describe the spatial usage characteristics of the research area, making up for the shortcomings of traditional research methods and providing relevant support for creating a good public space.

Keywords: computer vision, deep learning, public spaces, using features

Procedia PDF Downloads 36

Authors: Nuntaporn Aukkanit

Abstract:

The aim of this study was to determine the effect of oils on chemical, physical, and sensory properties of deep-fried dough stick. Five kinds of vegetable oil which were used for addition and frying consist of: palm oil, soybean oil, sunflower oil, rice bran oil, and canola oil. The results of this study showed that using different kinds of oil made significant difference in the quality of deep-fried dough stick. Deep-fried dough stick fried with the rice bran oil had the lowest moisture loss and oil absorption (p≤0.05), but it had some unsatisfactory physical properties (color, specific volume, density, and texture) and sensory characteristics. Nonetheless, deep-fried dough stick fried with the sunflower oil had moisture loss and oil absorption slightly more than the rice bran oil, but it had almost higher physical and sensory properties. Deep-fried dough sticks together with the sunflower oil did not have different sensory score from the palm oil, commonly used for production of deep-fried dough stick. These results indicated that addition and frying with the sunflower oil are appropriate for the production of deep-fried dough stick.

Keywords: deep-fried dough stick, palm oil, sunflower oil, rice bran oil

Procedia PDF Downloads 258

Authors: Luther Ollier, Sylvie Thiria, Anastase Charantonis, Carlos E. Mejia, Michel Crépon

Abstract:

Sea Surface Height Anomaly (SLA) is a signature of the sub-mesoscale dynamics of the upper ocean. Sea Surface Temperature (SST) is driven by these dynamics and can be used to improve the spatial interpolation of SLA fields. In this study, we focused on the temporal evolution of SLA fields. We explored the capacity of deep learning (DL) methods to predict short-term SLA fields using SST fields. We used simulated daily SLA and SST data from the Mercator Global Analysis and Forecasting System, with a resolution of (1/12)◦ in the North Atlantic Ocean (26.5-44.42◦N, -64.25–41.83◦E), covering the period from 1993 to 2019. Using a slightly modified image-to-image convolutional DL architecture, we demonstrated that SST is a relevant variable for controlling the SLA prediction. With a learning process inspired by the teaching-forcing method, we managed to improve the SLA forecast at five days by using the SST fields as additional information. We obtained predictions of a 12 cm (20 cm) error of SLA evolution for scales smaller than mesoscales and at time scales of 5 days (20 days), respectively. Moreover, the information provided by the SST allows us to limit the SLA error to 16 cm at 20 days when learning the trajectory.

Keywords: deep-learning, altimetry, sea surface temperature, forecast

Procedia PDF Downloads 54

Authors: Despoina Mouratidis, Katia Lida Kermanidis

Abstract:

Machine translation (MT) is a very important task in Natural Language Processing (NLP). MT evaluation is crucial in MT development, as it constitutes the means to assess the success of an MT system, and also helps improve its performance. Several methods have been proposed for the evaluation of (MT) systems. Some of the most popular ones in automatic MT evaluation are score-based, such as the BLEU score, and others are based on lexical similarity or syntactic similarity between the MT outputs and the reference involving higher-level information like part of speech tagging (POS). This paper presents a language-independent machine learning framework for classifying pairwise translations. This framework uses vector representations of two machine-produced translations, one from a statistical machine translation model (SMT) and one from a neural machine translation model (NMT). The vector representations consist of automatically extracted word embeddings and string-like language-independent features. These vector representations used as an input to a multi-layer neural network (NN) that models the similarity between each MT output and the reference, as well as between the two MT outputs. To evaluate the proposed approach, a professional translation and a "ground-truth" annotation are used. The parallel corpora used are English-Greek (EN-GR) and English-Italian (EN-IT), in the educational domain and of informal genres (video lecture subtitles, course forum text, etc.) that are difficult to be reliably translated. They have tested three basic deep learning (DL) architectures to this schema: (i) fully-connected dense, (ii) Convolutional Neural Network (CNN), and (iii) Long Short-Term Memory (LSTM). Experiments show that all tested architectures achieved better results when compared against those of some of the well-known basic approaches, such as Random Forest (RF) and Support Vector Machine (SVM). Better accuracy results are obtained when LSTM layers are used in our schema. In terms of a balance between the results, better accuracy results are obtained when dense layers are used. The reason for this is that the model correctly classifies more sentences of the minority class (SMT). For a more integrated analysis of the accuracy results, a qualitative linguistic analysis is carried out. In this context, problems have been identified about some figures of speech, as the metaphors, or about certain linguistic phenomena, such as per etymology: paronyms. It is quite interesting to find out why all the classifiers led to worse accuracy results in Italian as compared to Greek, taking into account that the linguistic features employed are language independent.

Keywords: machine learning, machine translation evaluation, neural network architecture, pairwise classification

Procedia PDF Downloads 103

Authors: Shimaa Holail, Tamer Saleh, Xiongwu Xiao

Abstract:

Building change detection (BCD) from satellite imagery is an essential topic in urbanization monitoring, agricultural land management, and updating geospatial databases. Recently, methods for detecting changes based on deep learning have made significant progress and impressive results. However, it has the problem of being insensitive to changes in buildings with complex spectral differences, and the features being extracted are not discriminatory enough, resulting in incomplete buildings and irregular boundaries. To overcome these problems, we propose a dual Siamese network based on the Unet model with the addition of a deep supervision strategy (DS) in this paper. This network consists of a backbone (encoder) based on ImageNet pre-training, a fusion block, and feature pyramid networks (FPN) to enhance the step-by-step information of the changing regions and obtain a more accurate BCD map. To train the proposed method, we created a new dataset (EGY-BCD) of high-resolution and multi-temporal aerial images captured over New Cairo in Egypt to detect building changes for this purpose. The experimental results showed that the proposed method is effective and performs well with the EGY-BCD dataset regarding the overall accuracy, F1-score, and mIoU, which were 91.6 %, 80.1 %, and 73.5 %, respectively.

Keywords: building change detection, deep supervision, semantic segmentation, EGY-BCD dataset

Procedia PDF Downloads 74

Authors: Beatriz Lafuente Alcázar, Yash Wani, Amit J. Nimunkar

Abstract:

Cardiovascular Diseases (CVDs) are the leading cause of death globally, and around 80% of sudden cardiac deaths are due to arrhythmias or irregular heartbeats. The majority of these pathologies are revealed by either short-term or long-term alterations in the electrocardiogram (ECG) morphology. The ECG is the main diagnostic tool in cardiology. It is a non-invasive, pain free procedure that measures the heart’s electrical activity and that allows the detecting of abnormal rhythms and underlying conditions. A cardiologist can diagnose a wide range of pathologies based on ECG’s form alterations, but the human interpretation is subjective and it is contingent to error. Moreover, ECG records can be quite prolonged in time, which can further complicate visual diagnosis, and deeply retard disease detection. In this context, deep learning methods have risen as a promising strategy to extract relevant features and eliminate individual subjectivity in ECG analysis. They facilitate the computation of large sets of data and can provide early and precise diagnoses. Therefore, the cardiology field is one of the areas that can most benefit from the implementation of deep learning algorithms. In the present study, a deep learning algorithm is trained following a novel approach, using a combination of different databases as the training set. The goal of the algorithm is to achieve the detection of R-peaks in ECG signals. Its performance is further evaluated in ECG signals with different origins and features to test the model’s ability to generalize its outcomes. Performance of the model for detection of R-peaks for clean and noisy ECGs is presented. The model is able to detect R-peaks in the presence of various types of noise, and when presented with data, it has not been trained. It is expected that this approach will increase the effectiveness and capacity of cardiologists to detect divergences in the normal cardiac activity of their patients.

Keywords: arrhythmia, deep learning, electrocardiogram, machine learning, R-peaks

Procedia PDF Downloads 143

Authors: Tingwei Shu, Dong Zhou, Chengjun Guo

Abstract:

Semantic communication is an emerging form of communication that realize intelligent communication by extracting semantic information of data at the source and transmitting it, and recovering the data at the receiving end. It can effectively solve the problem of data transmission under the situation of large data volume, low SNR and restricted bandwidth. With the development of Deep Learning, semantic communication further matures and is gradually applied in the fields of the Internet of Things, Uumanned Air Vehicle cluster communication, remote sensing scenarios, etc. We propose an improved semantic communication system for the situation where the data volume is huge and the spectrum resources are limited during the transmission of remote sensing images. At the transmitting, we need to extract the semantic information of remote sensing images, but there are some problems. The traditional semantic communication system based on Convolutional Neural Network cannot take into account the global semantic information and local semantic information of the image, which results in less-than-ideal image recovery at the receiving end. Therefore, we adopt the improved vision-Transformer-based structure as the semantic encoder instead of the mainstream one using CNN to extract the image semantic features. In this paper, we first perform pre-processing operations on remote sensing images to improve the resolution of the images in order to obtain images with more semantic information. We use wavelet transform to decompose the image into high-frequency and low-frequency components, perform bilinear interpolation on the high-frequency components and bicubic interpolation on the low-frequency components, and finally perform wavelet inverse transform to obtain the preprocessed image. We adopt the improved Vision-Transformer structure as the semantic coder to extract and transmit the semantic information of remote sensing images. The Vision-Transformer structure can better train the huge data volume and extract better image semantic features, and adopt the multi-layer self-attention mechanism to better capture the correlation between semantic features and reduce redundant features. Secondly, to improve the coding efficiency, we reduce the quadratic complexity of the self-attentive mechanism itself to linear so as to improve the image data processing speed of the model. We conducted experimental simulations on the RSOD dataset and compared the designed system with a semantic communication system based on CNN and image coding methods such as BGP and JPEG to verify that the method can effectively alleviate the problem of excessive data volume and improve the performance of image data communication.

Keywords: semantic communication, transformer, wavelet transform, data processing

Procedia PDF Downloads 47

Authors: Güzi̇n Ti̇rkeş, Özge Teki̇n, Kerem Kurtuluş, Y. Yekta Yurtseven, Murat Baran

Abstract:

A scanning electron microscope (SEM) is a type of electron microscope mainly used in nanoscience and nanotechnology areas. Automatic image recognition and classification are among the general areas of application concerning SEM. In line with these usages, the present paper proposes a deep learning algorithm that classifies SEM images into nine categories by means of an online application to simplify the process. The NFFA-EUROPE - 100% SEM data set, containing approximately 21,000 images, was used to train and test the algorithm at 80% and 20%, respectively. Validation was carried out using a separate data set obtained from the Middle East Technical University (METU) in Turkey. To increase the accuracy in the results, the Inception ResNet-V2 model was used in view of the Fine-Tuning approach. By using a confusion matrix, it was observed that the coated-surface category has a negative effect on the accuracy of the results since it contains other categories in the data set, thereby confusing the model when detecting category-specific patterns. For this reason, the coated-surface category was removed from the train data set, hence increasing accuracy by up to 96.5%.

Keywords: convolutional neural networks, deep learning, image classification, scanning electron microscope

Procedia PDF Downloads 84

Authors: M. K. Wijayarathna, S. M. Buddika Harshanath

Abstract:

Monitoring student engagement has become a crucial part of the educational process and a reliable indicator of the capacity to retain information. As online learning classrooms are now more common these days, students' attention levels have become increasingly important, making it more difficult to check each student's concentration level in an online classroom setting. To profile student attention to various gradients of engagement, a study is a plan to conduct using machine learning models. Using a convolutional neural network, the findings and confidence score of the high accuracy model are obtained. In this research, convolutional neural networks are using to help discover essential emotions that are critical in defining various levels of participation. Students' attention levels were shown to be influenced by emotions such as calm, enjoyment, surprise, and fear. An improved virtual learning system was created as a result of these data, which allowed teachers to focus their support and advise on those students who needed it. Student participation has formed as a crucial component of the learning technique and a consistent predictor of a student's capacity to retain material in the classroom. Convolutional neural networks have a plan to implement the platform. As a preliminary step, a video of the pupil would be taken. In the end, researchers used a convolutional neural network utilizing the Keras toolkit to take pictures of the recordings. Two convolutional neural network methods are planned to use to determine the pupils' attention level. Finally, those predicted student attention level results plan to display on the graphical user interface of the System.

Keywords: HTML5, JavaScript, Python flask framework, AI, graphical user

Procedia PDF Downloads 68

Authors: Mahdi Bazarganigilani

Abstract:

Aortic enlargement, also known as an aortic aneurysm, can occur when the walls of the aorta become weak. This disease can become deadly if overlooked and undiagnosed. In this paper, a computer-aided diagnosis (CAD) system was introduced to accurately diagnose aortic enlargement from chest x-ray images. An enhanced convolutional neural network (CNN) was employed and then trained by transfer learning by using three different main areas from the original images. The areas included the left lung, heart, and right lung. The accuracy of the system was then evaluated on 1001 samples by using 4-fold cross-validation. A promising accuracy of 90% was achieved in terms of the F-measure indicator. The results showed using different areas from the original image in the training phase of CNN could increase the accuracy of predictions. This encouraged the author to evaluate this method on a larger dataset and even on different CAD systems for further enhancement of this methodology.

Keywords: computer-aided diagnosis systems, aortic enlargement, chest X-ray, image processing, convolutional neural networks

Procedia PDF Downloads 127

Authors: Redouane Larbi Boufeniza, Jing-Jia Luo

Abstract:

This study assesses the suitability of deep learning (DL) for downscaling sea surface temperature (SST) over the Mediterranean Sea in the context of seasonal forecasting. We design a set of experiments that compare different DL configurations and deploy the best-performing architecture to downscale one-month lead forecasts of June–September (JJAS) SST from the Nanjing University of Information Science and Technology Climate Forecast System version 1.0 (NUIST-CFS1.0) for the period of 1982–2020. We have also introduced predictors over a larger area to include information about the main large-scale circulations that drive SST over the Mediterranean Sea region, which improves the downscaling results. Finally, we validate the raw model and downscaled forecasts in terms of both deterministic and probabilistic verification metrics, as well as their ability to reproduce the observed precipitation extreme and spell indicator indices. The results showed that the convolutional neural network (CNN)-based downscaling consistently improves the raw model forecasts, with lower bias and more accurate representations of the observed mean and extreme SST spatial patterns. Besides, the CNN-based downscaling yields a much more accurate forecast of extreme SST and spell indicators and reduces the significant relevant biases exhibited by the raw model predictions. Moreover, our results show that the CNN-based downscaling yields better skill scores than the raw model forecasts over most portions of the Mediterranean Sea. The results demonstrate the potential usefulness of CNN in downscaling seasonal SST predictions over the Mediterranean Sea, particularly in providing improved forecast products.

Keywords: Mediterranean Sea, sea surface temperature, seasonal forecasting, downscaling, deep learning

Procedia PDF Downloads 46

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 84

Authors: Yuntao Liu, Lei Wang, Haoran Xia

Abstract:

Machine learning has shown extensive applications in the development of classification models for autism spectrum disorder (ASD) using neural image data. This paper proposes a fusion multi-modal classification network based on a graph neural network. First, the brain is segmented into 116 regions of interest using a medical segmentation template (AAL, Anatomical Automatic Labeling). The image features of sMRI and the signal features of fMRI are extracted, which build the node and edge embedding representations of the brain map. Then, we construct a dynamically updated brain map neural network and propose a method based on a dynamic brain map adjacency matrix update mechanism and learnable graph to further improve the accuracy of autism diagnosis and recognition results. Based on the Autism Brain Imaging Data Exchange I dataset(ABIDE I), we reached a prediction accuracy of 74% between ASD and TD subjects. Besides, to study the biomarkers that can help doctors analyze diseases and interpretability, we used the features by extracting the top five maximum and minimum ROI weights. This work provides a meaningful way for brain disorder identification.

Keywords: autism spectrum disorder, brain map, supervised machine learning, graph network, multimodal data, model interpretability

Procedia PDF Downloads 20

Authors: Trupti Chavan, Ramachandra Guda, Kameshwar Rao

Abstract:

The lack of data is a pain problem in medical image analysis using a convolutional neural network (CNN). This work uses various spatial transformation techniques to address the medical image augmentation issue for knee detection and localization using an enhanced single shot detector (SSD) network. The spatial transforms like a negative, histogram equalization, power law, sharpening, averaging, gaussian blurring, etc. help to generate more samples, serve as pre-processing methods, and highlight the features of interest. The experimentation is done on the OpenKnee dataset which is a collection of knee images from the openly available online sources. The CNN called enhanced single shot detector (SSD) is utilized for the detection and localization of the knee joint from a given X-ray image. It is an enhanced version of the famous SSD network and is modified in such a way that it will reduce the number of prediction boxes at the output side. It consists of a classification network (VGGNET) and an auxiliary detection network. The performance is measured in mean average precision (mAP), and 99.96% mAP is achieved using the proposed enhanced SSD with spatial transformations. It is also seen that the localization boundary is comparatively more refined and closer to the ground truth in spatial augmentation and gives better detection and localization of knee joints.

Keywords: data augmentation, enhanced SSD, knee detection and localization, medical image analysis, openKnee, Spatial transformations

Procedia PDF Downloads 129

Authors: Bhaumik Tyagi, Mandeep Kaur, Kanika Singla

Abstract:

The advancement of blockchain has facilitated scholars to remodel e-voting systems for future generations. Server-side attacks like SQL injection attacks and DOS attacks are the most common attacks nowadays, where malicious codes are injected into the system through user input fields by illicit users, which leads to data leakage in the worst scenarios. Besides, quantum attacks are also there which manipulate the transactional data. In order to deal with all the above-mentioned attacks, integration of blockchain, convolutional neural network (CNN), and Quantum Key Distribution is done in this very research. The utilization of blockchain technology in e-voting applications is not a novel concept. But privacy and security issues are still there in a public and private blockchains. To solve this, the use of a hybrid blockchain is done in this research. This research proposed cryptographic signatures and blockchain algorithms to validate the origin and integrity of the votes. The convolutional neural network (CNN), a normalized version of the multilayer perceptron, is also applied in the system to analyze visual descriptions upon registration in a direction to enhance the privacy of voters and the e-voting system. Quantum Key Distribution is being implemented in order to secure a blockchain-based e-voting system from quantum attacks using quantum algorithms. Implementation of e-voting blockchain D-app and providing a proposed solution for the privacy of voters in e-voting using Blockchain, CNN, and Quantum Key Distribution is done.

Keywords: hybrid blockchain, secure e-voting system, convolutional neural networks, quantum key distribution, one-time pad

Procedia PDF Downloads 54

Authors: Harshit Mittal, Neeraj Garg

Abstract:

Hand symbol recognition is a pivotal component in the domain of computer vision, with far-reaching applications spanning sign language interpretation, human-computer interaction, and accessibility. This research paper discusses the approach with the integration of the Canny Edge algorithm and convolutional neural network. The significance of this study lies in its potential to enhance communication and accessibility for individuals with hearing impairments or those engaged in gesture-based interactions with technology. In the experiment mentioned, the data is manually collected by the authors from the webcam using Python codes, to increase the dataset augmentation, is applied to original images, which makes the model more compatible and advanced. Further, the dataset of about 6000 coloured images distributed equally in 5 classes (i.e., 1, 2, 3, 4, 5) are pre-processed first to gray images and then by the Canny Edge algorithm with threshold 1 and 2 as 150 each. After successful data building, this data is trained on the Convolutional Neural Network model, giving accuracy: 0.97834, precision: 0.97841, recall: 0.9783, and F1 score: 0.97832. For user purposes, a block of codes is built in Python to enable a window for hand symbol recognition. This research, at its core, seeks to advance the field of computer vision by providing an advanced perspective on hand sign recognition. By leveraging the capabilities of the Canny Edge algorithm and convolutional neural network, this study contributes to the ongoing efforts to create more accurate, efficient, and accessible solutions for individuals with diverse communication needs.

Keywords: hand symbol recognition, computer vision, Canny edge algorithm, convolutional neural network

Procedia PDF Downloads 34

Authors: Malik Muhammad Arslan, Muneeb Ullah, Dai Shihan, Daniyal Haider, Xiaodong Yang

Abstract:

Chest X-rays are instrumental in the detection and monitoring of a wide array of diseases, including viral infections such as COVID-19, tuberculosis, pneumonia, lung cancer, and various cardiac and pulmonary conditions. To enhance the accuracy of diagnosis, artificial intelligence (AI) algorithms, particularly deep learning models like Convolutional Neural Networks (CNNs), are employed. However, these deep learning models demand a substantial and varied dataset to attain optimal precision. Generative Adversarial Networks (GANs) can be employed to create new data, thereby supplementing the existing dataset and enhancing the accuracy of deep learning models. Nevertheless, GANs have their limitations, such as issues related to stability, convergence, and the ability to distinguish between authentic and fabricated data. In order to overcome these challenges and advance the detection and classification of CXR normal and abnormal images, this study introduces a distinctive technique known as DCWGAN (Diverse Conditional Wasserstein GAN) for generating synthetic chest X-ray (CXR) images. The study evaluates the effectiveness of this Idiosyncratic DCWGAN technique using the ResNet50 model and compares its results with those obtained using the traditional GAN approach. The findings reveal that the ResNet50 model trained on the DCWGAN-generated dataset outperformed the model trained on the classic GAN-generated dataset. Specifically, the ResNet50 model utilizing DCWGAN synthetic images achieved impressive performance metrics with an accuracy of 0.961, precision of 0.955, recall of 0.970, and F1-Measure of 0.963. These results indicate the promising potential for the early detection of diseases in CXR images using this Inimitable approach.

Keywords: CNN, classification, deep learning, GAN, Resnet50

Procedia PDF Downloads 47

Authors: Bubai Maji, Monorama Swain

Abstract:

Speech is considered an essential and most natural medium for the interaction between machines and humans. However, extracting effective features for speech emotion recognition (SER) is remains challenging. The present studies show that the temporal information captured but high-level temporal-feature learning is yet to be investigated. In this paper, we present an efficient novel method using the Self-attention (SA) mechanism in a combination of Convolutional Neural Network (CNN) and Bi-directional Gated Recurrent Unit (Bi-GRU) network to learn high-level temporal-feature. In order to further enhance the representation of the high-level temporal-feature, we integrate a Bi-GRU output with learnable weights features by SA, and improve the performance. We evaluate our proposed method on our created SITB-OSED and IEMOCAP databases. We report that the experimental results of our proposed method achieve state-of-the-art performance on both databases.

Keywords: Bi-GRU, 1D-CNNs, self-attention, speech emotion recognition

Procedia PDF Downloads 90