Deep Convolution Research Articles

Piecewise linear convolutional deep belief classifier for consumer Behavior Analysis in Social Network in Gaussian-embedded clustering

ABSTRACT A novel technique Gaussian-Embedded Clustering-based Piecewise Linear Convolutional Deep Belief Classifier (GEC-PLCDBC) is introduced for enhancing the accuracy of consumer behaviors pattern prediction with minimum time consumption. The proposed GEC-PLCDBC technique performs three major processes, namely, preprocessing, clustering, and behavior pattern analysis. First, preprocessing is carried out using Gaussian distributive neighbor embedding technique to remove the repeated, retweets, and duplicate tweets from the dataset. After preprocessing, the clustering process is said to be performed using Tversky qualitative Hartigan–Wong clustering for grouping similar tweets to minimize the time complexity of the behavior pattern analysis. Finally, the Piecewise Linear Regressive Convolutional Deep Belief Neural Classifier (PLRCDBNC) is used for behavioral pattern prediction such as positive, negative, or neutral based on the extracted words from the tweets. In this way, the GEC-PLCDBC technique improves behavioral pattern prediction while maintaining a low error rate. GEC-PLCDBC technique’s experimental assessment is carried out with respect to clustering accuracy, behavior pattern prediction accuracy, behavior pattern prediction time, and error rate, with respect to different number of tweets. The quantitative results show that GEC-PLCDBC outperforms existing methods by 8% in clustering accuracy and 8% in prediction accuracy of behavior pattern prediction with 15% less time and 26% mean absolute error than the conventional methods.

CNN-MRI Detection of Fatty Infiltration, Rotator Cuff and Infraspinatus Muscle Atrophy in Shoulder Pain Patients

Half of the global population has or has had shoulder fractures due to routine activities, whether intentional or unintentional. Designing a user-friendly program to input Shoulder fracture images into the model and execute it is a significant problem in the area of Shoulder Fracture Diagnosis Research. The machine is capable of predicting shoulder fractures, but it is unable to anticipate the specific subtypes of fractures that may result from other occurrences or disorders. This research aims to create an innovative Deep Convolution Neural Networks system for accurately predicting the diagnosis of shoulder fractures, in order to facilitate appropriate treatment. Our goal is to provide these services to assist individuals or groups in overcoming the delayed detection of shoulder fractures and eliminating the need for physical diagnosis of shoulder fractures in hospitals. The objective of designing a unique categorizing system is to provide a collection of test case inputs that ensure comprehensive coverage over the test area at a predetermined level. This yields a collection of test cases that prioritize the execution of the feature without considering the specific implementation details.

A Snapshot Multi-Spectral Demosaicing Method for Multi-Spectral Filter Array Images Based on Channel Attention Network

Multi-spectral imaging technologies have made great progress in the past few decades. The development of snapshot cameras equipped with a specific multi-spectral filter array (MSFA) allow dynamic scenes to be captured on a miniaturized platform across multiple spectral bands, opening up extensive applications in quantitative and visualized analysis. However, a snapshot camera based on MSFA captures a single band per pixel; thus, the other spectral band components of pixels are all missed. The raw images, which are captured by snapshot multi-spectral imaging systems, require a reconstruction procedure called demosaicing to estimate a fully defined multi-spectral image (MSI). With increasing spectral bands, the challenge of demosaicing becomes more difficult. Furthermore, the existing demosaicing methods will produce adverse artifacts and aliasing because of the adverse effects of spatial interpolation and the inadequacy of the number of layers in the network structure. In this paper, a novel multi-spectral demosaicing method based on a deep convolution neural network (CNN) is proposed for the reconstruction of full-resolution multi-spectral images from raw MSFA-based spectral mosaic images. The CNN is integrated with the channel attention mechanism to protect important channel features. We verify the merits of the proposed method using 5 × 5 raw mosaic images on synthetic as well as real-world data. The experimental results show that the proposed method outperforms the existing demosaicing methods in terms of spatial details and spectral fidelity.

Non-intrusive, transferable model for coupled turbulent channel-porous media flow based upon neural networks

Turbulent flow over permeable interfaces is omnipresent featuring complex flow topology. In this work, a data-driven, end-to-end machine learning model has been developed to model the turbulent flow in porous media. For the same, we have derived a non-linear reduced order model (ROM) with a deep convolution autoencoder. This model can reduce highly resolved spatial dimensions, which is a prerequisite for direct numerical simulation, by 99%. A downstream recurrent neural network has been trained to capture the temporal trend of reduced modes; thus, it is able to provide future evolution of modes. We further evaluate the trained model's capability on a newer dataset with a different porosity. In such cases, fine-tuning could reduce the efforts (up to two-order of magnitude) to train a model with limited dataset (10%) and knowledge and still show a good agreement on the mean velocity profile. Especially, the fine-tuned model shows a better agreement in the porous domain than the channel and interface areas indicating the topological feature is less challenging for training than the multi-scale nature of the turbulent flows. Leveraging the current model, we find that even quick fine-tuning achieves an impressive order-of-magnitude reduction in training time by approximately O(102) and still results in effective flow predictions. This promising discovery encourages the fast development of a substantial amount of data-driven models tailored for various types of porous media. The diminished training time substantially lowers the computational cost when dealing with changing porous topologies, making it feasible to systematically explore interface engineering with different types of porous media.

Human Skin Disease Detection Using Machine Learning

Abstract: In this research, I want to improve deep convolution neural networks that have been successful with the ImageNet dataset at categorizing seven different types of skin lesions using the HAM10000 dataset, which has 10,000 dermatoscopic pictures. With VGG16, Inception V3, Inception ResNet V2, and Dense Net 201, the top layers were fine-tuned.

UnifiedTT: Visual tracking with unified transformer

Target tracking is an important research task in computer vision. Existing tracking algorithms based on Siamese networks often suffer from the problem of information redundancy between adjacent frames and a lack of ability to capture global dependencies. When similar backgrounds appear around the target, the tracking performance usually significantly decreases. Although target tracking algorithms based on deep convolution and the Transformer have partially addressed these issues, achieving a good balance between the two remains a challenge. In this work, we propose a unified convolution and self-attention Siamese network for target tracking. By utilizing a feature extraction backbone network based on integrated convolution and self-attention styles, we are able to capture globally important regions and key frames while greatly reducing local redundant computations, thereby improving tracking performance. We apply this approach to the task of target tracking and enhance the feature extraction capabilities of both the target template and search template. Experimental results show that our proposed tracking algorithm outperforms some recent classical tracking algorithms, especially achieving improvements of 10.7 % on the high-diversity dataset GOT-10K and 24.7 % on the large-scale and high-quality LaSOT dataset.

A comparative analysis of CNN-based deep learning architectures for early diagnosis of bone cancer using CT images

Bone cancer is a rare in which cells in the bone grow out of control, resulting in destroying the normal bone tissue. A benign type of bone cancer is harmless and does not spread to other body parts, whereas a malignant type can spread to other body parts and might be harmful. According to Cancer Research UK (2021), the survival rate for patients with bone cancer is 40% and early detection can increase the chances of survival by providing treatment at the initial stages. Prior detection of these lumps or masses can reduce the risk of death and treat bone cancer early. The goal of this current study is to utilize image processing techniques and deep learning-based Convolution neural network (CNN) to classify normal and cancerous bone images. Medical image processing techniques, like pre-processing (e.g., median filter), K-means clustering segmentation, and, canny edge detection were used to detect the cancer region in Computer Tomography (CT) images for parosteal osteosarcoma, enchondroma and osteochondroma types of bone cancer. After segmentation, the normal and cancerous affected images were classified using various existing CNN-based models. The results revealed that AlexNet model showed a better performance with a training accuracy of 98%, validation accuracy of 98%, and testing accuracy of 100%.

Contrastive deep convolutional transform k-means clustering

Deep clustering has gained the immense attention of researchers in recent years. Most of the deep clustering approaches are based on auto-encoders which consist of an encoder-decoder framework. In these approaches, the clustering module is embedded in the latent space of auto-encoders. The auto-encoder based deep clustering approaches require learning of encoder weights as well as decoder weights. Moreover, due to the unsupervised learning strategy, these approaches lack in learning the discriminative features that can help in generating better clusters. This work introduces a novel clustering approach based on Contrastive Deep Convolutional Transform Learning (DCTL) framework. The proposed approach mitigates the problem of lack of supervision in DCTL based K-means clustering approach by embedding the contrastive learning into it. To embed the contrastive learning, the positive pairs and negative pairs of data samples are generated by reconstructing the data samples from the DCTL learnt representation itself and thus eliminates the requirement of data augmentation for embedding contrastive learning. The experimental results on several benchmark facial images datasets demonstrate that the proposed framework gives better clustering performance as compared to the current state-of-the-art deep clustering approaches especially in data constrained scenarios.

Method to enhance time series rolling fault prediction by deep fast Fourier convolution

Deep learning realizes the adaptive extraction of features in remaining useful life (RUL), and most of the methods combined with traditional signal processing stay in the preprocessing stage. A fast Fourier convolutional gated recurrent unit (FFCGRU) method is proposed, which incorporates fast Fourier transform (FFT) into a convolutional neural network for adaptive feature extraction, enhancing fault prediction. First, a plug-and-play FFT convolution block is proposed. This method combines signal processing with the adaptive deep learning process and achieves a larger receptive field through a smaller convolution kernel. Secondly, an adaptive pooling layer is designed before the feature map is input into the gated recurrent unit. The feature maps of different channels are aggregated into one feature value to reduce the redundant information of the signal feature map. Comparative studies indicate that the FFCGRU method utilizes a small number of parameters even with large convolutional kernels and exhibits robustness.

Taylor African vulture optimization algorithm with hybrid deep convolution neural network for image captioning system

Taylor African vulture optimization algorithm with hybrid deep convolution neural network for image captioning system

Gear fault diagnosis based on complex network theory and error-correcting output codes: Multi class support vector machine

Gearbox failures have a detrimental effect on the machine performance that affects the production capacity and economic benefits of a manufacturing industry in an adverse manner. Early detection of faults in gears helps to prevent sudden machine failures. Previously, many signal processing and artificial intelligence techniques have been successfully applied to diagnose gear faults by analyzing the machine vibrations. These techniques have repeatedly emphasized on the need to extract quality features from gear vibration signals and using artificial intelligence techniques that can identify multiple types of faults simultaneously. As such, in this article, new features called as visibility graph features based on complex network theory have been proposed to extract fault information from vibration signals. In addition, a new artificial intelligence framework for multifault classification known as error-correcting output codes-multiclass support vector machine is proposed to detect gear faults. To the best of authors’ knowledge, these features along with the error-correcting output codes-artificial intelligence framework have been rarely utilized for fault diagnosis purpose. The proposed approach has been applied on experimental gear vibration data to validate its effectiveness. First, the vibration signals are transformed to graphs, and various graph properties are computed to serve as fault features. Then, the fault features are supplied to train the error-correcting output codes-multiclass support vector machine model and learn the fault patterns. Finally, the gear fault classification accuracies are determined for various gear test conditions and compared with those of existing methods. It is observed that the proposed approach provides an average improvement of 4.22%, 9.93% and 11.8% over the time-domain features, time–frequency domain features and voting-based multiclass support vector machine classifier, respectively, for different operating conditions. Furthermore, the suggested technique augments the classification accuracies by 10%, 7.5% and 6.7%, respectively, as compared with the deep-learning models, namely standard stacked sparse autoencoder, deep neural network and convolution neural network.

Integrated Generative Adversarial Networks and Deep Convolutional Neural Networks for Image Data Classification: A Case Study for COVID-19

Convolutional Neural Networks (CNNs) have garnered significant utilisation within automated image classification systems. CNNs possess the ability to leverage the spatial and temporal correlations inherent in a dataset. This study delves into the use of cutting-edge deep learning for precise image data classification, focusing on overcoming the difficulties brought on by the COVID-19 pandemic. In order to improve the accuracy and robustness of COVID-19 image classification, the study introduces a novel methodology that combines the strength of Deep Convolutional Neural Networks (DCNNs) and Generative Adversarial Networks (GANs). This proposed study helps to mitigate the lack of labelled coronavirus (COVID-19) images, which has been a standard limitation in related research, and improves the model’s ability to distinguish between COVID-19-related patterns and healthy lung images. The study uses a thorough case study and uses a sizable dataset of chest X-ray images covering COVID-19 cases, other respiratory conditions, and healthy lung conditions. The integrated model outperforms conventional DCNN-based techniques in terms of classification accuracy after being trained on this dataset. To address the issues of an unbalanced dataset, GAN will produce synthetic pictures and extract deep features from every image. A thorough understanding of the model’s performance in real-world scenarios is also provided by the study’s meticulous evaluation of the model’s performance using a variety of metrics, including accuracy, precision, recall, and F1-score.

Electrolytic capacitor surface defect detection based on deep convolution neural network

The existing methods for detecting surface defects in electrolytic capacitors are typically based on conventional machine vision, with limited feature extraction capabilities, poor versatility, slow detection speed, and the inability to achieve accurate and real-time defect detection. In this study, a real-time object detection algorithm based on an improved single shot multibox detector (SSD) is proposed to achieve omnidirectional surface defect detection of electrolytic capacitors. First, an electrolytic capacitor surface image acquisition device was established to capture omnidirectional surface images of the capacitors, and an electrolytic capacitor surface defect dataset was created. Next, the visual geometry group (VGG)-16 network structure was replaced with the MobileNetv2 network structure, effectively reducing the model’s parameter count and improving inference speed. Moreover, the Multibox Loss function was replaced with the Focal Loss function to increase the model’s attention to difficult-to-classify samples and improve model accuracy. Additionally, a transfer learning network model was designed to apply the model to electrolytic capacitors of different colors using small sample learning. Finally, the performance of the improved network model was tested on a dataset of electrolytic capacitor surface defects. The experimental results demonstrate that the parameters quantity of improved model is 3.50 M, the mAP value reaches 92.67 %, which is improved by 2.54 %, and the Macro-F1 value reaches 92.15 %, which is 11.32 % higher than that before improvement. Thus, the proposed improved SSD model provides a theoretical basis and technical prerequisites for automated and intelligent surface defect detection in electrolytic capacitors.

A novel approach for bearings multiclass fault diagnosis fusing multiscale deep convolution and hybrid attention networks

Insufficient and imbalanced samples pose a significant challenge in bearing fault diagnosis, leading to low diagnosis accuracy. However, the fault characteristics of vibration signals are weak and difficult to extract when faults occur in the early stage. This paper proposes an effective fault diagnosis method that addresses small and imbalanced sample problems under noise interference. First, the number of faulty samples in the form of 1D signals is increased mainly by the sliding split sampling method. The preprocessed data are used to create 2D time–frequency diagrams using the continuous wavelet transform (CWT), which can extract effective features to improve the data quality. Subsequently, the minority samples are oversampled by combining synthetic minority oversampling technique to realize time–frequency conversion augmented oversampling. Moreover, the clustering method and random undersampling method are introduced to prevent the overfitting and underfitting problems respectively. Then, we propose a hybrid attention mechanism to enhance the extraction of effective feature information. This combination, integrating CWT with a multicolumn modified deep residual network, effectively extracts fault characteristics and suppresses noise effects. The experimental results demonstrate the effectiveness of the proposed method by comparison with other advanced methods using two case studies of bearing datasets.

Feature similarity learning based on fuzziness minimization for semi-supervised medical image segmentation

Deep learning has advanced the automation and intelligence levels of medical image segmentation, but the acquisition of annotations for medical images proves to be very challenging. Deep semi-supervised learning provides an effective approach to tackling this challenge. In this paper, we propose a method of feature similarity learning for deep semi-supervised image segmentation. The method performs a weighted transformation on the latent features of deep convolution network, and the objective of this transformation is designed to minimize the uncertainty of feature similarity. Based on fuzzy sets of similarity, we define the fuzziness to measure the uncertainty in the objective, and demonstrate that the optimization objective is equivalent to strengthening the discriminant property of those transformed features. This similarity learning is independent of the pseudo-labels, boosting the robustness against noisy pseudo-labels. Our method is implemented through a plug-and-play neural network layer, which can be optimized alongside the segmentation model in an end-to-end manner. The model validation focuses on the left atrium segmentation which is a challenging task for atrial fibrillation treatment. Comprehensive experiments demonstrate the effectiveness of the uncertainty-guided similarity learning on semi-supervised segmentation tasks.

Deep convolutional dictionary learning network for sparse view CT reconstruction with a group sparse prior

Purpose Numerous techniques based on deep learning have been utilized in sparse view computed tomography (CT) imaging. Nevertheless, the majority of techniques are instinctively constructed utilizing state-of-the-art opaque convolutional neural networks (CNNs) and lack interpretability. Moreover, CNNs tend to focus on local receptive fields and neglect nonlocal self-similarity prior information. Obtaining diagnostically valuable images from sparsely sampled projections is a challenging and ill-posed task.Method To address this issue, we propose a unique and understandable model named DCDL-GS for sparse view CT imaging. This model relies on a network comprised of convolutional dictionary learning and a nonlocal group sparse prior. To enhance the quality of image reconstruction, we utilize a neural network in conjunction with a statistical iterative reconstruction framework and perform a set number of iterations. Inspired by group sparsity priors, we adopt a novel group thresholding operation to improve the feature representation and constraint ability and obtain a theoretical interpretation. Furthermore, our DCDL-GS model incorporates filtered backprojection (FBP) reconstruction, fast sliding window nonlocal self-similarity operations, and a lightweight and interpretable convolutional dictionary learning network to enhance the applicability of the model.Results The efficiency of our proposed DCDL-GS model in preserving edges and recovering features is demonstrated by the visual results obtained on the LDCT-P and UIH datasets. Compared to the results of the most advanced techniques, the quantitative results are enhanced, with increases of 0.6-0.8 dB for the peak signal-to-noise ratio (PSNR), 0.005-0.01 for the structural similarity index measure (SSIM), and 1-1.3 for the regulated Fréchet inception distance (rFID) on the test dataset. The quantitative results also show the effectiveness of our proposed deep convolution iterative reconstruction module and nonlocal group sparse prior.Conclusion In this paper, we create a consolidated and enhanced mathematical model by integrating projection data and prior knowledge of images into a deep iterative model. The model is more practical and interpretable than existing approaches. The results from the experiment show that the proposed model performs well in comparison to the others.

Downscaling Seasonal Precipitation Forecasts over East Africa with Deep Convolutional Neural Networks

Downscaling Seasonal Precipitation Forecasts over East Africa with Deep Convolutional Neural Networks

YOLO_Bolt: a lightweight network model for bolt detection

Accurate, fast, and intelligent workpiece identification is of great significance to industrial production. To cope with the limited hardware resources of factory equipment, we have made lightweight improvements based on You Only Look Once v5 (YOLOv5) and proposed a lightweight YOLO named YOLO_Bolt. First, ghost bottleneck lightweight deep convolution is added to the backbone module and neck module of the YOLOv5 detection algorithm to reduce the model volume. Second, the asymptotic feature pyramid network is added to enhance the feature utilization ability, suppress interference information, and improve detection accuracy. Finally, the relationship between the loss function and the decoupling head structure was focused on, and the number of decoupling head layers was redesigned according to different tasks to further improve the detection accuracy of the workpiece detection model. We conducted experimental verification on the MSCOCO 2017 dataset and the homemade bolt dataset. The experimental results show that compared with YOLOv5s, the number of model parameters is only 6.8 M, which is half that of the original model. On the MSCOCO 2017 dataset, the mAP increased by 2.4%. FPS increased by 104 frames/s. On the homemade dataset, the mAP 0.5 increased by 4.2%, and our proposed method is 1.2% higher than the latest YOLOv8s. The improved network can provide effective auxiliary technical support for workpiece detection.

Fishing risky behavior recognition based on adaptive transformer, reinforcement learning and stochastic configuration networks

The fishing behavior in a high-voltage environment may lead to electric shock accidents. This paper proposes a fishing risky behavior recognition method based on adaptive Transformer, reinforcement learning and stochastic configuration networks (SCNs). Firstly, the fishing behavior image sample training set is evaluated using the actor network based on linear entropy in the reinforcement learning module to select suitable training samples that possess abundant straight-line elements. Subsequently, the selected training samples are fed into a multi-scale spatial deep convolution to extract continuous spatial features of elongated objects. The deformable Transformer network, enhanced with adaptive encoding and decoding layers through SCNs, is used to obtain the position and detection results of the fishing rod. The trained Transformer model is evaluated by a defined credibility evaluation metric for acquiring rewards to update the training sample set iteratively. Then, an adaptive adjustment mechanism is constructed for the encoding and decoding layers of the adaptive Transformer network to establish a library of adaptive Transformer models with different encoding and decoding layer levels to accommodate the feature requirements of training samples in various scenarios. Finally, the blending learning method for combining the detection results from the model library is integrated with human body object detection and pose estimation methods, and a predefined expert system is utilized for logic reasoning on fishing risky behavior. Experimental results demonstrate the effectiveness of the proposed method in this paper.

IMCNN:Intelligent Malware Classification using Deep Convolution Neural Networks as Transfer learning and ensemble learning in honeypot enabled organizational network

Traditional malware detection systems based on signature-based detection methods cannot detect new and unseen malware. Moreover, conventional machine learning methods for malware detection have utilized features extracted through static program analysis or dynamic analysis, which requires code debugging and execution primarily through offline processing; hence not a scalable approach. This paper proposes a novel intelligent malware classification using a deep convolution neural network (IMCNN) in organizational networks enabled with Honeypots. Systematic customization of pre-trained convolutional neural networks(CNN) as a transfer learning and ensemble learning as a classification is presented to detect intelligent modern-day malware. Real-world malware samples are systematically labeled and visualized into grayscale images. Four cutting-edge deep CNN models - VGG16, VGG19, InceptionV3, and ResNet50, are trained on the ImageNet database (≥1 million) and fine-tuned as feature extractors along with a basic CNN model. Three strategies are designed for feature extraction and selection: Rectified linear unit (ReLU) fully connected layer embedded in a deep CNN model, principal component analysis (PCA), and singular value decomposition(SVD). Reduced sets of features are stacked and used to train k-nearest neighbor (k-NN), support vector machine (SVM), and random forest (RF) classifiers for predictions. Subsequently, the predictive probabilities of different machine-learned models are ensembled using a soft voting method for final classification. The proposed method is evaluated on MalImg datasets (9339 malware samples of 25 families) and real-world modern malware datasets (690 malware of 22 families). The experimental results reveal that despite using a reduced feature set, the IMCNN effectively detects malware with 99.36% test accuracy on unseen data for MalImg datasets and 92.11% for real-world malware. In addition, the proposed method is compared with several existing state-of-art malware detection models in terms of performance accuracy and found performing as the best. Experiments demonstrated that the proposed method is resilient to polymorphic code obfuscation used by the malware authors.