Variant Of Convolutional Neural Network Research Articles

Automatic classification of multi-carrier modulation signal using STFT spectrogram and deep CNN

In the realm of communication systems, categorizing Multi-Carrier Modulation (MCM) signals without cooperative communication poses a significant technical challenge. In this paper, we introduce a novel approach for accurately categorizing five distinct MCM signals, including Orthogonal Frequency Division Multiplexing (OFDM), Filter Bank Multicarrier (FBMC), Filtered Orthogonal Frequency Division Multiplexing (FOFDM), Windowed Orthogonal Frequency Division Multiplexing (WOLA), and Universal Filtered Multicarrier (UFMC). Each signal is considered with two types of subcarrier waveforms, Quadrature Amplitude Modulation 16 (QAM16) and Quadrature Amplitude Modulation 64 (QAM64), resulting in a total of 10 unique MCM signals for classification. Our proposed methodology leverages Short-Time Fourier Transform (STFT) spectrograms for feature extraction at the frontend, while at the backend, we employ three variants of Convolutional Neural Network (CNN) models; CNN, CNN with Dropout (CNN_d), CNN with both Dropout and L1 Regularization (CNN_dL1) and one deep CNN model; Xception, individually. We aim to provide an efficient and reliable means of categorizing MCM signals, with practical applications in signal processing and communication systems. Extensive simulations demonstrate the effectiveness of our approach, achieving remarkable accuracies. Notably, the Xception model exhibits the highest accuracy among the four models considered. Specifically, we attain an accuracy of 98% at 10 dB SNR using the Xception model. These results underscore the efficacy of our proposed methodology and highlight the potential for its deployment in real-world scenarios.

Emergency sign language recognition from variant of convolutional neural network (CNN) and long short term memory (LSTM) models

Sign language is the primary communication tool used by the deaf community and people with speaking difficulties, especially during emergencies. Numerous deep learning models have been proposed to solve the sign language recognition problem. Recently. Bidirectional LSTM (BLSTM) has been proposed and used in replacement of Long Short-Term Memory (LSTM) as it may improve learning long-team dependencies as well as increase the accuracy of the model. However, there needs to be more comparison for the performance of LSTM and BLSTM in LRCN model architecture in sign language interpretation applications. Therefore, this study focused on the dense analysis of the LRCN model, including 1) training the CNN from scratch and 2) modeling with pre-trained CNN, VGG-19, and ResNet50. Other than that, the ConvLSTM model, a special variant of LSTM designed for video input, has also been modeled and compared with the LRCN in representing emergency sign language recognition. Within LRCN variants, the performance of a small CNN network was compared with pre-trained VGG-19 and ResNet50V2. A dataset of emergency Indian Sign Language with eight classes is used to train the models. The model with the best performance is the VGG-19 + LSTM model, with a testing accuracy of 96.39%. Small LRCN networks, which are 5 CNN subunits + LSTM and 4 CNN subunits + BLSTM, have 95.18% testing accuracy. This performance is on par with our best-proposed model, VGG + LSTM. By incorporating bidirectional LSTM (BLSTM) into deep learning models, the ability to understand long-term dependencies can be improved. This can enhance accuracy in reading sign language, leading to more effective communication during emergencies.

Orchard Mapping with Deep Learning Semantic Segmentation

Abstract: The goal of orchard mapping with deep learning semantic segmentation is to automatic detection and localization of the orchard tree canopy under varied situations like multiple seasons, various tree ages, and varying levels of weed covering. The accuracy of segmentation depends on many factors like season, canopy size, and presence of weeds, background soil condition, and untreated soil. The proposed research have several combinations of training & test data based on orchard conditions The proposed research work contains deep learning convolutional neural network variant. The algorithm U- net network will be used. This helps to automatic detect and localize the tree canopy size using UAV images. Image segmentation will help all farmers by automatically segmenting tree canopies from aerial images for the development of integrated tools. After measurement of these parameters, tree canopy volume can be detected of targeted tree

Lane Detection and Traffic Sign Detection using Deep Learning and Computer Vision for Autonomous Driving Research Using CARLA Simulator

Lane identification and traffic sign detection is the most challenging and promising problem for self-driving or autonomous vehicles with unintentional lane departure and ignorance of traffic signs being major contributing factors to motor vehicle collisions around the world. To tackle this problem the proposed work aims to detect both lane and traffic signs for autonomous vehicles. This article proposes semantic segmentation and object detection model for implementing Advanced Driver Assistance System (ADAS) applications. The applications are implemented using a variant of Convolutional Neural Networks (CNN) deep learning model such as SegNet and You Only Look Once (YOLO) algorithm. Due to dynamic and adverse environment conditions, devising and testing a system which yields effective performance in all urban driving scenarios is challenging. Hence the environment is set up virtually with the help of the Car Learning to Act (CARLA) simulator. With aid of developed models, lane and traffic signs were successfully detected and tested under various constraints. Obtained results are evaluated with various performance metrics. Models are deployed for separate created datasets. The semantic segmentation model developed for lane detection using SegNet gives a mean average precision (mAP) of 93.33%, an overall accuracy of 94.80%, F-score of 93.42% and a minimal error rate of 5.20%. Model developed for Traffic sign detection, a mean average precision of 93.67%, an accuracy of 95.56%, recall of 92.67%, F-score of 93.16% and error rate of 4.44% have been achieved.

Enhanced capsule network‐based executable files malware detection and classification—deep learning approach

SummaryMalware has considerably increased recently, posing a serious security danger to both people and enterprises. In order to distinguish and stop the negative effects of malware, a variety of machine and deep learning approaches have been used to detect it. However, while extracting malware features, the feature‐to‐feature spatial hierarchy is not taken into account by the existing techniques and as a result, information is lost during the pooling operation. Hence, a modified capsule deep neural network was developed in which discriminative features are extracted from three channel image derived from malware binary with considering feature‐to‐feature spatial hierarchy. Also, conventional capsule deep neural network is modified by adding a global average pooling layer before fully connected layer thereby classified the dataset as malicious or benign without any loss of information. Moreover, these malwares were not accurately classified based on their families using existing variants of convolutional neural network (CNN) since malware family variants can modify due to minute changes in malware binaries. Hence, a hybrid deep convolutional neural network (DCNN) and long‐short‐term memory (LSTM) has been utilized that determine minute changes in malware binaries using LSTM without vanishing gradient issue and effectively perform malware family classification using DCNN. As a result, the proposed approach successfully identifies malware in executable files and categorizes malware into families with 98.5% accuracy.

Multi-layer capsule network with joint dynamic routing for fire recognition

Fire recognition has emerged as a critical concern in the field of fire safety. Deep learning techniques, specifically convolutional neural networks (CNNs), have found widespread application in fire recognition tasks. The capsule network, a higher-level variant of CNN, offers distinct advantages in terms of enhanced recognition accuracy, making it suitable for fire recognition applications. However, the capsule network faces challenges in effectively determining the presence or absence of fire objects due to the idiosyncrasies of its dynamic routing algorithm. To address this issue and enable effective fire recognition, we propose a novel approach that involves a multi-layer capsule network. Within this framework, we introduce a joint dynamic routing algorithm that computes output values during forward propagation within the multi-layer capsule network. Additionally, we present a new loss function and a fully connected auxiliary training layer designed to train the multi-layer capsule networks effectively. Comparative evaluations against conventional CNN architectures and recent state-of-the-art fire recognition methods demonstrate that the enhanced multi-layer capsule network achieves higher test accuracy, even with limited training samples and fewer training iterations. Furthermore, owing to its reduced model parameter scale, the multi-layer capsule network exhibits faster recognition speeds compared to the aforementioned methods.

CDBMGCIG: Design of a Cross-Domain Bioinspired Model for identification of Gait Components via Iterated GANs

This Gait identification assists in recognition of human body components from temporal image sequences. Such components consist of connected-body entities including head, upper body, lower body regions. Existing Gait recognition models use deep learning methods including variants of Convolutional Neural Networks (CNNs), Q-Learning, etc. But these methods are either highly complex, or do not perform well under complex background conditions. Moreover, most of these models are validated on a specific environmental condition, and cannot be scaled for general-purpose deployments. To overcome these issues, this text proposes design of a novel cross-domain bioinspired model for identification of gait components via Iterated Generative Adversarial Networks (IGANs). The proposed model initially extracts multidomain pixel-level feature sets from different images. These include frequency components via Fourier analysis, entropy components via Cosine analysis, spatial components via Gabor analysis, and window-based components via Wavelet &Convolutional analysis. These feature sets are processed via a Grey Wolf Optimization (GWO) Model, which assists in identification of high-density & highly variant features for different gait components. These features are classified via an iterated GAN, which comprises of Generator & Discriminator ssModels that assist in evaluating connected body components. These operations generate component-level scores that assist in identification of gait from complex background images. Due to which, the proposed model was observed to achieve 9.5% higher accuracy, 3.4% higher precision, and 2.9% higher recall than existing gait identification methods. The model also uses iterative learning, due to which its accuracy is incrementally improved w.r.t. number of evaluated image sets.

Enhancing Fingerprint Liveness Detection Accuracy Using Deep Learning: A Comprehensive Study and Novel Approach.

Liveness detection for fingerprint impressions plays a role in the meaningful prevention of any unauthorized activity or phishing attempt. The accessibility of unique individual identification has increased the popularity of biometrics. Deep learning with computer vision has proven remarkable results in image classification, detection, and many others. The proposed methodology relies on an attention model and ResNet convolutions. Spatial attention (SA) and channel attention (CA) models were used sequentially to enhance feature learning. A three-fold sequential attention model is used along with five convolution learning layers. The method's performances have been tested across different pooling strategies, such as Max, Average, and Stochastic, over the LivDet-2021 dataset. Comparisons against different state-of-the-art variants of Convolutional Neural Networks, such as DenseNet121, VGG19, InceptionV3, and conventional ResNet50, have been carried out. In particular, tests have been aimed at assessing ResNet34 and ResNet50 models on feature extraction by further enhancing the sequential attention model. A Multilayer Perceptron (MLP) classifier used alongside a fully connected layer returns the ultimate prediction of the entire stack. Finally, the proposed method is also evaluated on feature extraction with and without attention models for ResNet and considering different pooling strategies.

Pixel‐level thin crack detection on road surface using convolutional neural network for severely imbalanced data

AbstractBecause roads are the major backbone of the transportation network, research about crack detection on road surfaces has been popular in computer and infrastructure engineering. When training a convolutional neural network (CNN) for pixel‐level road crack detection, three common challenges include (1) the data are severely imbalanced, (2) crack pixels can be easily confused with normal road texture and other visual noises, and (3) there are many unexplainable characteristics regarding the CNN itself. When it comes to very fine and thin cracks, these challenges are exaggerated and a new challenge is introduced, as there can be a discrepancy between the actual width and the annotated width of a crack. To tackle all these challenges of thin crack detection, this paper proposes a new variant of CNN named ThinCrack U‐Net, designed to provide thin results upon pixel‐level crack detection on road surfaces. The main contribution is to demystify how pixel‐level thin crack detection results are affected by different loss functions as well as various combinations of the U‐Net components. The experimental results show that ThinCrack U‐Net yields a significant performance boost in CrackTree260, from 65.71% to 94.48% F‐measure, compared to the existing variant of U‐Net previously proposed in the context of pixel‐level thin crack detection. Finally, this paper locates the source of undesirable result thickness and solves it with the balanced usage of downsampling/upsampling layers and atrous convolution. Unlike suggested by previous works, different loss functions show no significant impact on ThinCrack U‐Net, whereas normalization layers are proved crucial in pixel‐level thin crack detection.

Toward trustworthy human suspicious activity detection from surveillance videos using deep learning

AbstractIn today’s world, suspicious or unusual activities express threats and danger to others. For the prevention of various security issues, an automatic video detection system is very important. It is difficult to consecutively monitor camera videos recorded in public places to detect any abnormal event, so an automated video detection system is needed. The study objective is to create an intelligent and trustworthy system that will take a video stream as input and detect what kind of suspicious activity is happening in that video to reduce the time that consumes watching the video. In this work, we use three models Convolutional Neural Network (CNN), GRU, and ConvLSTM model. These models are trained on the same dataset of 6 suspicious activities of humans that are: Running, Punching, Falling, Snatching, Kicking, and Shooting. The dataset consists of various videos related to each activity. Different deep learning techniques are applied in the proposed work: preprocessing, data annotation model training, and classification. The frames are extracted from the source video, and then features are calculated through the model in Inception v3, a Convolutional Neural Network variant. On the same dataset, the CNN model attains 91.55% accuracy, the ConvLSTM model attains 88.73% accuracy, and the GRU model attains 84.01% accuracy. The performance of the proposed models is evaluated using a confusion matrix, F1-Score, precision, and recall. The proposed model proved better than other models in terms of performance and accuracy. The findings of this study prove helpful unusual events by examining a person's abnormal behavior.

Polarformer: Optic Disc and Cup Segmentation Using a Hybrid CNN-Transformer and Polar Transformation

The segmentation of optic disc (OD) and optic cup (OC) are used in the automatic diagnosis of glaucoma. However, the spatially ambiguous boundary and semantically uncertain region-of-interest area in pictures may lead to the degradation of the performance of precise segmentation of the OC and OD. Unlike most existing methods, including the variants of CNNs (Convolutional Neural Networks) and U-Net, which limit the contributions of rich global features, we instead propose a hybrid CNN-transformer and polar transformation network, dubbed as Polarformer, which aims to extract discriminative and semantic features for robust OD and OC segmentation. Our Polarformer typically exploits contextualized features among all input units and models the correlation of structural relationships under the paradigm of the transformer backbone. More specifically, our learnable polar transformer module optimizes the polar transformations by sampling images in the Cartesian space and then mapping them back to the polar coordinate system for masked-image reconstruction. Extensive experimental results present that our Polarformer achieves superior performance in comparison to most state-of-the-art methods on three publicly available datasets.

FPGA-Based Acceleration for Bayesian Convolutional Neural Networks

Neural networks (NNs) have demonstrated their potential in a variety of domains ranging from computer vision (CV) to natural language processing. Among various NNs, two-dimensional (2-D) and three-dimensional (3-D) convolutional NNs (CNNs) have been widely adopted for a broad spectrum of applications, such as image classification and video recognition, due to their excellent capabilities in extracting 2-D and 3-D features. However, standard 2-D and 3-D CNNs are not able to capture their model uncertainty which is crucial for many safety-critical applications, including healthcare and autonomous driving. In contrast, Bayesian CNNs (BayesCNNs), as a variant of CNNs, have demonstrated their ability to express uncertainty in their prediction via a mathematical grounding. Nevertheless, BayesCNNs have not been widely used in industrial practice due to their compute requirements stemming from sampling and subsequent forward passes through the whole network multiple times. As a result, these requirements significantly increase the amount of computation and memory consumption in comparison to standard CNNs. This article proposes a novel field-programmable gate array (FPGA)-based hardware architecture to accelerate both 2-D and 3-D BayesCNNs based on Monte Carlo dropout (MCD). Compared with other state-of-the-art accelerators for BayesCNNs, the proposed design can achieve up to four times higher energy efficiency and nine times better compute efficiency. An automatic framework capable of supporting partial Bayesian inference is proposed to explore the tradeoff between algorithm and hardware performance. Extensive experiments are conducted to demonstrate that our framework can effectively find the optimal implementations in the design space.

A Survey of Electric Load Forecasting Algorithm Models

With the increasing economic and social development, the load forecasting of the power system plays an increasingly important role in many energy-related fields such as energy dispatching, market-based price forecasting, and capacity allocation for the government and power companies. Based on the long-term research of many scholars, this paper aims to provide a general overview of some specific categories of load forecasting methods and models. Linear models can meet the real needs of forecasting in the early days, represented by autoregressive moving average (ARIMA) , but the lack of modeling ability requires researchers to develop more novel models based on machine learning and deep learning. Algorithms are widely used in the field of load forecasting, including human neural network (ANN) , support vector machine (SVM) , various variants of convolutional neural network (CNN) , memory network (RNN , LSTM...) , etc., which enhance the prediction performance and each have their own unique advantages. While the depth of the network continues to deepen, residual networks are proposed to optimize the model and solve a series of intractable problems. Different models are integrated and integrated together to form a new integrated network to give full play to the advantages of each model. However, these models do not solve the problem once and for all and have their own limitations. Finally, three different experiments are used to compare the three types of models mentioned in this article, and some typical network models are used as examples for performance analysis and demonstration. This review paper summarizes the valuable information of the prediction network, provides valuable information for subsequent research, and provides perspectives and entry points for subsequent research work.

Fault Detection Utilizing Convolution Neural Network on Timeseries Synchrophasor Data From Phasor Measurement Units

An end-to-end supervised learning method is proposed for fault detection in the electric grid using Big Data from multiple Phasor Measurement Units (PMUs). The approach consists of preprocessing steps aimed at reducing data noise and dimensionality, followed by utilization of six classification models considered for detecting faults. Three of the models were variants of Convolutional Neural Network (CNN) architectures that consider a single type of measurement (voltage, current or frequency) at all PMUs or all types together also at all PMUs. CNN based models were compared to traditional methods of Logistic Regression (LR), Multi-layer Perceptron (MLP) and Support Vector Machine (SVM). Evaluation was conducted on two-year data measured by PMUs at 37 locations in a large electric grid. The response variable for classification were extracted from the grid-wide outage event log. Experiments show that CNN-based models outperformed traditional methods on one year out-of-sample outage detection over the entire grid.

A Bayesian deep learning method for freeway incident detection with uncertainty quantification

Incident detection is fundamental for freeway management to reduce non-recurrent congestions and secondary incidents. Recently, machine learning technologies have made considerable progress in the incident detection field, but many still face challenges in uncertainty quantification due to the aleatoric uncertainty of traffic data and the epistemic uncertainty of model deviations. In this study, a Bayesian deep learning method was proposed for freeway incident detection with uncertainty quantification. A convolutional neural network variant was designed on a Bayesian framework, and mechanisms of Bayes by backpropagation and local reparameterization technics were used to update the weight of the proposed model. The predictive uncertainty of the proposed method was modeled jointly by integrating the aleatoric and epistemic uncertainty. The proposed model was tested on the PORTAL dataset and compared with four benchmark models: standard normal deviate, wavelet neural network, long-short term memory neural network, and convolutional neural network. The results show that the proposed model outperforms the baseline methods in terms of accuracy, detection rate and false alarm rate. Perturbation experiments were used to test the robustness of the model against noise. The results indicated that the aleatoric uncertainty of the model remained almost constant under different noise levels. The proposed method may benefit future studies on uncertainty quantification while using machine learning method in incident management and other fields in intelligent transportation systems.

Multi-Lesion Segmentation of Diabetic Retinopathy Using Deep Learning

Abstract: Diabetic Retinopathy (DR) and Diabetic Macular Edema (DME) are the two major complications of diabetes and have a significant impact on working individuals of the world population. DR doesn’t give any early symptoms. Therefore, it is important to diagnose DR at an early stage. The two above mentioned diseases usually depend on the presence and areas of lesions in fundus images. The four main related lesions include soft exudates, hard exudates,microaneurysms, and haemorrhages. Since lesions in retinal fundus images are a pivotal indicator of DR, analyzing retinal fundus images is the most popular method for DR screening. The examination of fundus images is time-consuming and small lesions are hard to observe. Therefore, adopting deep learning techniques for lesion segmentation is of great importance. In this project, we use one of the deep learning techniques called U-Net, which is a variant of Convolutional Neural Networks (CNN) for multiple lesion segmentation.

A Comprehensive Survey of Image-Based Food Recognition and Volume Estimation Methods for Dietary Assessment.

Dietary studies showed that dietary problems such as obesity are associated with other chronic diseases, including hypertension, irregular blood sugar levels, and increased risk of heart attacks. The primary cause of these problems is poor lifestyle choices and unhealthy dietary habits, which are manageable using interactive mHealth apps. However, traditional dietary monitoring systems using manual food logging suffer from imprecision, underreporting, time consumption, and low adherence. Recent dietary monitoring systems tackle these challenges by automatic assessment of dietary intake through machine learning methods. This survey discusses the best-performing methodologies that have been developed so far for automatic food recognition and volume estimation. Firstly, the paper presented the rationale of visual-based methods for food recognition. Then, the core of the study is the presentation, discussion, and evaluation of these methods based on popular food image databases. In this context, this study discusses the mobile applications that are implementing these methods for automatic food logging. Our findings indicate that around 66.7% of surveyed studies use visual features from deep neural networks for food recognition. Similarly, all surveyed studies employed a variant of convolutional neural networks (CNN) for ingredient recognition due to recent research interest. Finally, this survey ends with a discussion of potential applications of food image analysis, existing research gaps, and open issues of this research area. Learning from unlabeled image datasets in an unsupervised manner, catastrophic forgetting during continual learning, and improving model transparency using explainable AI are potential areas of interest for future studies.

Comparative Analysis of Deep Convolutional Neural Network Models for Humerus Bone Fracture Detection

Deep learning plays a key role in medical image processing. One of the applications of deep learning models in this domain is bone fracture detection from X-ray images. Convolutional neural network and its variants are used in wide range of medical image processing applications. MURA Dataset is commonly used in various studies that detect bone fractures and this work also uses that dataset, in specific the Humerus bone radiograph images. The humerus dataset in the MURA dataset contains both images with fracture and without fracture. The image with fracture includes images with metals which are removed in this work. Experimental analysis was made with two variants of convolutional neural network, DenseNet169 Model and the VGG Model. In case of the DenseNet169 model, a model with the pre trained weights of ImageNet and one without it is experimented. Results obtained with these variants of CNN are comparedand it shows that DenseNet169 model that uses pre-trained weights of ImageNet model performs better than the other two models.

Classification of nutrient deficiencies in rice crop using denseNet-BC

The symptoms associated with deficiencies in plants tends to appear often on the leaves. The color and shape of a leaf often used for diagnosing the nutritional deficiencies in plants and classification of these properties often pose serious problem. Since same color and shape may have many root cause problem. It is hence necessary to carefully analyse the texture of leaf with proper training of a classifier. In this paper, we design an image acquisition based classification model that utilizes Internet of Things (IoTs) for data acquisition and a variant of convolutional neural networks (CNN) namely DenseNet-BC for the task of classification. Prior classification, the model is trained over several iteration based on careful observation of features and its related symptoms. The simulation is conducted with fine-tuning of classification after several iterations. The results of simulation shows that the proposed method obtains improved classification accuracy in terms of accuracy and F-measure than other deep learning models.

Improved Deep Convolutional Neural Network to Classify Osteoarthritis from Anterior Cruciate Ligament Tear Using Magnetic Resonance Imaging.

Anterior cruciate ligament (ACL) tear is caused by partially or completely torn ACL ligament in the knee, especially in sportsmen. There is a need to classify the ACL tear before it fully ruptures to avoid osteoarthritis. This research aims to identify ACL tears automatically and efficiently with a deep learning approach. A dataset was gathered, consisting of 917 knee magnetic resonance images (MRI) from Clinical Hospital Centre Rijeka, Croatia. The dataset we used consists of three classes: non-injured, partial tears, and fully ruptured knee MRI. The study compares and evaluates two variants of convolutional neural networks (CNN). We first tested the standard CNN model of five layers and then a customized CNN model of eleven layers. Eight different hyper-parameters were adjusted and tested on both variants. Our customized CNN model showed good results after a 25% random split using RMSprop and a learning rate of 0.001. The average evaluations are measured by accuracy, precision, sensitivity, specificity, and F1-score in the case of the standard CNN using the Adam optimizer with a learning rate of 0.001, i.e., 96.3%, 95%, 96%, 96.9%, and 95.6%, respectively. In the case of the customized CNN model, using the same evaluation measures, the model performed at 98.6%, 98%, 98%, 98.5%, and 98%, respectively, using an RMSprop optimizer with a learning rate of 0.001. Moreover, we also present our results on the receiver operating curve and area under the curve (ROC AUC). The customized CNN model with the Adam optimizer and a learning rate of 0.001 achieved 0.99 over three classes was highest among all. The model showed good results overall, and in the future, we can improve it to apply other CNN architectures to detect and segment other ligament parts like meniscus and cartilages.