Convolutional Deep Belief Network Research Articles

Comprehensive Quality Evaluation Model of College Students Based on Deep Learning

Evaluating college students' quality in higher education institutions is significant for promoting enhancement and innovation in teaching and learning practices. Conventional evaluation strategies mainly depend on subjective evaluation, which limits their ability to capture the multifaceted characteristics of students. In this study, we developed a distinct deep learning (DL) architecture by combining the Convolutional Deep Belief Network and Red Fox Optimization (CDBN-RFO) to evaluate college students' quality. Initially, a database containing academic performance records, extracurricular activities, and other relevant information is collected and fed into the system. Then, the database was preprocessed to make it an appropriate format for subsequent analysis. Further, the proposed CDBN-RFO was created, and the CDBN was trained using the preprocessed database to understand the patterns and relations within data for evaluating college students' quality. Subsequently, the RFO approach was deployed to fine-tune the CDBN hyperparameters to their finest range, which increases the overall system performance. Subsequently, the RFO optimized the CDBN training process by fine-tuning its parameters to the optimal range, enhancing overall evaluation performance. The developed framework was modeled using Python software, and the results were analyzed, including accuracy, error, computational time, etc. Also, a comparative analysis was done with conventional algorithms, which validated that the proposed strategy outperformed them in accuracy.

Convolutional Deep Belief Network Based Expert System for Automated Fault Diagnosis in Hydro Electrical Power Systems

The paper developed an approach for fault diagnosis in Hydro-Electrical Power Systems (HEPS). Using a Renewable Energy System (RES), HEPS has performed a significant part in contributing to addressing the evolving energy demands of the present. Several electro-mechanical elements that collectively comprise the Hydro-Electric (HE) system are susceptible to corrosion from routine usage and unplanned occurrences. Administration and servicing systems that are successful in implementing and achieving these goals are those that regularly track and predict failures. Detect models applied in the past included those that were primarily reactive or reliant on human involvement to identify and analyse abnormalities. The significant multiple variables intricacies that impact successful fault detection are disregarded by these frameworks. The research presented here proposes a Convolutional Deep Belief Network (CDBN) driven Deep Learning (DL) model for successful fault and failure detection in such power systems that address these problems. Applying sample data collected from two Chinese power plants, the proposed framework has been assessed compared to other practical DL algorithms. Different metrics have been employed to determine the effectiveness of the simulations, namely Accuracy, Precision, Recall, and F1-score. These outcomes indicated that the CDBN is capable of predicting unexpected failures. Graphic representations demonstrating control used to measure turbine blade load, vibration level, and generator heat for assessing the replicas.

3D object recognition using deep learning for automatically generating semantic BIM data

The successful reuse of Building Information Model (BIM) data is reliant on the use of clearly defined objects. File formats such as the Industry Foundation Classes (IFC) along with classification systems offer approaches to standardising semantic BIM data. Inconsistent application of these standards during BIM authoring results in objects that are unable to be reused across systems. This research proposes a Machine Learning (ML) based alternative for detecting objects and enriching BIM data based on geometrical representations of objects to support interoperability across systems throughout the supply chain. First, an existing dataset of training and test data is modified to correspond with predefined data types from the IFC schema. Second, a convolutional deep belief network is implemented and applied on new testing data from the National BIM Library to identify 3D objects and enabling semantic enrichment of BIM data. Finally, the testing is extended to four real-world industry BIM models. The findings show that the geometric data from emerging BIM object libraries can be used to search and retrieve objects within industry BIM models based on geometric rather than semantic data.

A deep learning fusion model for accurate classification of brain tumours in Magnetic Resonance images

AbstractDetecting brain tumours is complex due to the natural variation in their location, shape, and intensity in images. While having accurate detection and segmentation of brain tumours would be beneficial, current methods still need to solve this problem despite the numerous available approaches. Precise analysis of Magnetic Resonance Imaging (MRI) is crucial for detecting, segmenting, and classifying brain tumours in medical diagnostics. Magnetic Resonance Imaging is a vital component in medical diagnosis, and it requires precise, efficient, careful, efficient, and reliable image analysis techniques. The authors developed a Deep Learning (DL) fusion model to classify brain tumours reliably. Deep Learning models require large amounts of training data to achieve good results, so the researchers utilised data augmentation techniques to increase the dataset size for training models. VGG16, ResNet50, and convolutional deep belief networks networks extracted deep features from MRI images. Softmax was used as the classifier, and the training set was supplemented with intentionally created MRI images of brain tumours in addition to the genuine ones. The features of two DL models were combined in the proposed model to generate a fusion model, which significantly increased classification accuracy. An openly accessible dataset from the internet was used to test the model's performance, and the experimental results showed that the proposed fusion model achieved a classification accuracy of 98.98%. Finally, the results were compared with existing methods, and the proposed model outperformed them significantly.

An Enhanced Detection Method of PCB Defect Based on D-DenseNet (PCBDD-DDNet)

Printed Circuit Boards (PCBs), as integral components of electronic products, play a crucial role in modern industrial production. However, due to the precision and complexity of PCBs, existing PCB defect detection methods exhibit some issues such as low detection accuracy and limited usability. In order to address these problems, a PCB defect detection method based on D-DenseNet (PCBDD-DDNet) has been proposed. This method capitalizes on the advantages of two deep learning networks, CDBN (Convolutional Deep Belief Networks) and DenseNet (Densely Connected Convolutional Networks), to construct the D-DenseNet (Combination of CDBN and DenseNet) network. Within this network, CDBN focuses on extracting low-level features, while DenseNet is responsible for high-level feature extraction. The outputs from both networks are integrated using a weighted averaging approach. Additionally, the D-DenseNet employs a multi-scale module to extract features from different levels. This is achieved by incorporating filters of sizes 3 × 3, 5 × 5, and 7 × 7 along the three paths of the CDBN network, multi-scale feature extraction network, and DenseNet network, effectively capturing information at various scales. To prevent overfitting and enhance network performance, the Adafactor optimization function and L2 regularization are introduced. Finally, online hard example mining mechanism (OHEM) is incorporated to improve the network’s handling of challenging samples and enhance the accuracy of the PCB defect detection network. The effectiveness of this PCBDD-DDNet method is demonstrated through experiments conducted on publicly available PCB datasets. And the method achieves a mAP (mean Average Precision) of 93.24%, with an accuracy higher than other classical networks. The results affirm the method’s efficacy in PCB defect detection.

Microphone Array Speech Enhancement Via Beamforming Based Deep Learning Network

In general, in-car speech enhancement is an application of the microphone array speech enhancement in particular acoustic environments. Speech enhancement inside the moving cars is always an interesting topic and the researchers work to create some modules to increase the quality of speech and intelligibility of speech in cars. The passenger dialogue inside the car, the sound of other equipment, and a wide range of interference effects are major challenges in the task of speech separation in-car environment. To overcome this issue, a novel Beamforming based Deep learning Network (Bf-DLN) has been proposed for speech enhancement. Initially, the captured microphone array signals are pre-processed using an Adaptive beamforming technique named Least Constrained Minimum Variance (LCMV). Consequently, the proposed method uses a time-frequency representation to transform the pre-processed data into an image. The smoothed pseudo-Wigner-Ville distribution (SPWVD) is used for converting time-domain speech inputs into images. Convolutional deep belief network (CDBN) is used to extract the most pertinent features from these transformed images. Enhanced Elephant Heard Algorithm (EEHA) is used for selecting the desired source by eliminating the interference source. The experimental result demonstrates the effectiveness of the proposed strategy in removing background noise from the original speech signal. The proposed strategy outperforms existing methods in terms of PESQ, STOI, SSNRI, and SNR. The PESQ of the proposed Bf-DLN has a maximum PESQ of 1.98, whereas existing models like Two-stage Bi-LSTM has 1.82, DNN-C has 1.75 and GCN has 1.68 respectively. The PESQ of the proposed method is 1.75%, 3.15%, and 4.22% better than the existing GCN, DNN-C, and Bi-LSTM techniques. The efficacy of the proposed method is then validated by experiments.

SASG-GCN: Self-Attention Similarity Guided Graph Convolutional Network for Multi-Type Lower-Grade Glioma Classification

Identifying the subtypes of low-grade glioma (LGG) can help prevent brain tumor progression and patient death. However, the complicated non-linear relationship and high dimensionality of 3D brain MRI limit the performance of machine learning methods. Therefore, it is important to develop a classification method that can overcome these limitations. This study proposes a self-attention similarity-guided graph convolutional network (SASG-GCN) that uses the constructed graphs to complete multi-classification (tumor-free (TF), WG, and TMG). In the pipeline of SASG-GCN, we use a convolutional deep belief network and a self-attention similarity-based method to construct the vertices and edges of the constructed graphs at 3D MRI level, respectively. The multi-classification experiment is performed in a two-layer GCN model. SASG-GCN is trained and evaluated on 402 3D MRI images which are produced from the TCGA-LGG dataset. Empirical tests demonstrate that SASGGCN accurately classifies the subtypes of LGG. The accuracy of SASG-GCN achieves 93.62%, outperforming several other state-of-the-art classification methods. In-depth discussion and analysis reveal that the self-attention similarity-guided strategy improves the performance of SASG-GCN. The visualization revealed differences between different gliomas.

Intelligent Diagnosis Using Continuous Wavelet Transform and Gauss Convolutional Deep Belief Network

Bearing fault diagnosis is of significance to ensure the safe and reliable operation of a motor. Deep learning provides a powerful ability to extract the features of raw data automatically. A convolutional deep belief network (CDBN) is an effective deep learning method. In this article, a novel vibration amplitude spectrum imaging feature extraction method using continuous wavelet transform and image conversion is proposed, which can extract the image features with two-dimensional and eliminate the effect of handcrafted features under low signal-to-noise ratio conditions, different operating conditions, and data segmentation. Then, a novel CDBN with Gaussian distribution is constructed to learn the representative features for bearing fault classification. The proposed method is tested on motor bearing dataset with four and ten classifications. The results have been compared with other methods. The experiment results show that the proposed method has achieved significant improvements and is more effective than the traditional methods.

Emotion classification using EEG signal for women safety application based on deep learning

An emotion is a conscious logical response that varies for different situations in women’s life. These mental responses are caused by physiological, cognitive, and behavioral changes. Gender-based violence undermines the participation of women in decision-making, resulting in a decline in their quality of life. More accurate and automatic classification of women’s emotions can enhance human-computer interfaces and security in real time. There are some wearable technologies and mobile applications that claim to ensure the safety of women. However, they rely on limited social action and are ineffective at ensuring women’s safety when and where it is needed. In this work, a novel CDB-LSTM network has been proposed to accurately classify the emotions of women in seven different classes. The electroencephalogram (EEG) offers non-radioactive methods of identifying emotions. Initially, the EEG signals are preprocessed and they are converted into images via Time-Frequency Representation (TPR). A smoothed pseudo-Wigner-Ville distribution (SPWVD) is employed to convert the EEG time-domain signals into input images. Consequently, these converted images are given as input to the Convolutional Deep Belief Network (CDBN) for extracting the most relevant features. Finally, Bi-directional LSTM is used for classifying the emotions of women into seven classes namely: happy, relax, sad, fear, anxiety, anger, and stress. The proposed CDB-LSTM network preserves the high accuracy range of 97.27% in the validation phase. The proposed CDB-LSTM network improves the overall accuracy by 6.20% 32.98% 6.85% and 3.30% better than CNN-LSTM, Multi-domain feature fusion model, GCNN-LSTM and CNN with SVM and DT respectively.

Linear feature projective geometric damped convolutional deep belief network for indoor floor planning

Wireless localization or positioning is essential for delivering location-based services for designing location tracking systems. Traditional indoor floor planning system employs wireless signals for accurate position estimation. But these positioning schemes failed to perform position estimation effectively and accurately through many obstacles or objects. The novel technique called Linear Features Projective Geometric Damped Convolutional Deep Belief Network (LFPGDCDBN) is introduced to improve the position estimation accuracy with minimum error. The proposed LFPGDCDBN technique includes two major processes namely dimensionality reduction and position estimation. First, the dimensionality reduction process is performed by projecting the principle features using Linear Helmert–Wolf blocked Sammon projection. After the feature selection, Geometric Levenberg–Marquardt Convolutional deep belief network is employed to estimate the position of the devices with higher accuracy and minimum error. The Convolutional deep belief network uses the triangulation geometric method to identify the position of the device in an indoor positioning system. Then the Levenberg–Marquardt function is a Damped least square method to minimize the squares of the deviations between the expected and observed results at the output unit. As a result, the LFPGDCDBN increases the positioning accuracy and minimizes the error rate. Experimental MATLAB assessment is carried out with various factors such as computational time, Computational space, positioning accuracy, and positing error. The experimental results and discussion indicate that the proposed LFPGDCDBN provides improved performance in terms of achieving higher positioning accuracy and minimum error as well as computational time when compared to the existing methods. The experimental results and discussion indicate that the proposed LFPGDCDBN increases the positioning accuracy by 47% and computational time, computational complexity, and reduces the positioning error by 45%, 29%, and 74% as compared to state-of-the-art works.

Preface

The 2022 International Conference on Intelligent Equipment and Data Processing (IEDP 2022) was successfully held on October 21st-23rd, 2022 in Guangzhou, China (virtual conference). IEDP 2022 encouraged exchange of information on frontiers of research in different areas, connected the most advanced academic resources in China and the world, and turned research results into industrial solutions.We had the honor of having invited Prof. Guixiong Liu from South China University of Technology, China to serve as our Conference General Chair. 90 individuals participated in the conference. Divided into three parts, the conference agenda covered keynote speeches, oral presentations, and online Q&A discussion. Firstly, keynote speakers were each allocated 30-45 minutes to hold their speeches. Then in the oral presentations, the excellent papers we had selected were presented by their authors one by one.During the conference, we were honored to invite four sophisticated keynote speakers to address their speeches. Among them, Prof. Wenbing Zhao from Cleveland State University, America delivered a speech on Video-Based Baseball Pitch Type Recognition. During the speech, he reported a preliminary study on video-based pitch type recognition using deep learning. A semi-automated way of building datasets based on publicly available video and pitch information for MLB games was proposed to facilitate the study and the two-stream inflated 3D convolutional neural network (I3D) was used for pitch type recognition. Next, Prof. Ming Jiang from Sun Yat-sen University, China addressed a presentation on Massive MIMO Beam Vector Assignment Based on Deep Learning and Location/Handover Information. In this paper, he proposed to apply the Gaussian anomaly detection (GAD) to improve the system performance, particularly in the handover zone, by exploiting the information of sea lane and locations of the user equipment (UE) and used convolutional deep belief networks (CDBN) and GAD in the further analysis. Their insightful speeches had triggered heated discussion in the third part of the conference. Besides, every participant praised this conference for disseminating useful and insightful knowledge.The proceedings of IEDP 2022 are a compilation of the accepted papers and represent an interesting outcome of the conference. These papers feature but are not limited to the following topics: Intelligent Inspection Technology, Hybrid Intelligent Systems, Integrated Data Terminals, Data Science and Big Data Technology, etc. All the papers have been checked through a rigorous review and process to meet the requirements of publication.We would like to acknowledge all of those who supported IEDP 2022 and made it a great success. Particularly, we would like to thank the Journal of Physics: Conference Series, for the efforts of all its colleagues in publishing this paper volume. All the help and contribution of each individual and institution was instrumental in the success of IEDP 2022.The Committee of IEDP 2022List of Committee member is available in this pdf.

Deep learning aided cyclostationary feature analysis for blind modulation recognition in massive MIMO systems

Blind modulation recognition (BMR) has been proposed as a promising approach for massive multiple-input multiple-output (M-MIMO) systems to support massive user equipment (UE). In this work, we devise a new uplink BMR algorithm based on a deep learning (DL) aided cyclostationary feature (CF) analysis approach. Firstly, we propose the so-called minimum description length (MDL) method aided complex fast independent component analysis (CFICA) algorithm, to separate the multi-user (MU) signals subjected to arbitrary, different modulation schemes with an unknown number of transmitters. Then, we tailor a convolutional deep belief network (CDBN) to analyze the relation between CF and modulation type (MT) for generating the BMR results. As the proposed BMR process is invoked blindly, it is particularly suitable for non-cooperative MU communication scenarios. Simulation results demonstrate the effectiveness of the proposed technique.

Convolutional Deep Belief Network Based Short Text Classification on Arabic Corpus

With a population of 440 million, Arabic language users form the rapidly growing language group on the web in terms of the number of Internet users. 11 million monthly Twitter users were active and posted nearly 27.4 million tweets every day. In order to develop a classification system for the Arabic language there comes a need of understanding the syntactic framework of the words thereby manipulating and representing the words for making their classification effective. In this view, this article introduces a Dolphin Swarm Optimization with Convolutional Deep Belief Network for Short Text Classification (DSOCDBN-STC) model on Arabic Corpus. The presented DSOCDBN-STC model majorly aims to classify Arabic short text in social media. The presented DSOCDBN-STC model encompasses preprocessing and word2vec word embedding at the preliminary stage. Besides, the DSOCDBN-STC model involves CDBN based classification model for Arabic short text. At last, the DSO technique can be exploited for optimal modification of the hyperparameters related to the CDBN method. To establish the enhanced performance of the DSOCDBN-STC model, a wide range of simulations have been performed. The simulation results confirmed the supremacy of the DSOCDBN-STC model over existing models with improved accuracy of 99.26%.

Special issue on International Conference on Computing and Communication Networks

Developments in computer communications and networks have led to exciting new areas of research and application, including the internet of things, vehicular networks, collaborative big data analysis, to name just a few. Moreover, the design and implementation of energy efficient future generation communication and networking technologies have fostered the development of new mobile, pervasive, and large-scale computing technologies. The International Conference on Computing and Communications Networks serves as a forum for exchanging the latest findings and experiences ranging from theoretical research to practical system development in all aspects of computing and networking. In its 2021 edition (ICCCN2021), many of those contributions had a significant artificial intelligence component, which is the focus of this Special Issue, which brings together selected papers from ICCCN2021. Of around 25 submitted articles, only 10 papers were selected based on their reviews. Each paper was reviewed by at least two reviewers and went through at least two rounds of reviews. The contributions of these papers are summarized briefly below. Ibrahim et al., 2022 propose a developed system to create a reliable COVID-19 prediction network using various layers starting with the segmentation of the lung CT scan image and ending with disease prediction. The initial step of the system starts with a proposed technique for lung segmentation that relies on a no-threshold histogram-based image segmentation method. Afterward, the GrabCut method was used as a post-segmentation method to enhance segmentation outcomes and avoid over- and under-segmentation problems. Then, three pre-trained models of standard DL methods, including Visual Geometry Group Network, convolutional deep belief network, and high-resolution network, were utilized to extract the most affective features from the segmented images that can help to identify COVID-19. Gupta et al., 2022 present a new AI technique based on optimal deep convolutional neural network (AI-ODCNN) for retinal fundus image classification. Primarily, the proposed model uses the Gaussian Blur based noise removal and contrast enhancement technique (CLAHE) based contrast enhancement technique to pre-process the retinal fundus image. In addition, morphology and contour-based image segmentation is performed. Moreover, the deep CNN with RMSProp Optimizer is employed for retinal fundus image classification. A wide range of simulations was performed on the automated retinal image analysis and structured analysis of the retina and the outcomes are examined with respect to various measures. The simulation outcomes ensured the better performance of the proposed approach related to other recent algorithms with maximum accuracy of 96.47%. Zhang et al., 2022 present a genetic algorithm-based on-orbit self-repair implementation for SRAM FPGAs to address the disadvantages of the genetic algorithm-based fault repair method in the aerospace industry. The proposed method has been verified by various applications in XC7VX330T, which demonstrates its engineering practicability. Mishra & Kohli, 2022 present the Anas platyrhynchos optimizer with deep learning-enabled block-based motion estimation (APODL-BBME) model. The proposed model estimates motion using block-based concepts and DL approaches. To accomplish this, the training and testing frames are separated into non-overlapping block categories, and the blocks are filtered via the bilateral filtering (BF) approach. In addition, a histogram of gradients (HOG) and densely connected network (DenseNet) model are employed to extract features, which are then fed into the bidirectional long short-term memory (BiLSTM) model to classify the input features. Finally, the APO algorithm is applied to optimally tune the hyperparameters of the BiLSTM model, which helps to improve the overall motion estimation efficacy, showing the novelty of the work. To demonstrate the enhanced performance of the APODL-BBME model, a comprehensive analysis is carried out, and when we compare the results, the APODL-BBME model outperforms recent motion estimation approaches. Bedi et al., 2022 propose an unsupervised approach for extractive summarization, based on semantic similarity and keyword-phrase extraction. Merging Concept Map and the RAKE method, a generic summary is computed based on threshold values. Both single-document and multi-document summarization can be accomplished with the approach. To evaluate the proposed unsupervised approach, various biomedical transcripts of neuro-science, general medicine, gastroenterology, orthopaedic and radiology domain are used. MT Sample Dataset is used to collect 1040 different transcripts. Using the proposed approach, an average ROUGE score of 0.77 for single-document summarization; however, for generic summary, an average ROUGE is 0.72. The proposed technique is validated for the previous corpus of BioMed articles, and results are better with state-of-the-art techniques. Marzouk et al., 2022 introduce a Quasi-Oppositional Wild Horse Optimization-based Multi-Agent Path Finding (QOWHO-MAPF) scheme for real-time IoT systems. The aim of the proposed QOWHO-MAPF scheme is to determine the optimal set of paths to reach the destination in real-time IoT networks. QOWHO algorithm is created by integrating the concepts of Quasi-Oppositional Based Learning (QOBL) and conventional WHO algorithm. In addition, the proposed QOWHO-MAPF model derives a fitness function that involves two input parameters such as residual energy and distance-to-destination. The proposed QOWHO-MAPF model was experimentally analysed and the results were inspected under several aspects. The simulation results established that QOWHO-MAPF model is a superior model compared with other state-of-the-art models. Vinoth & Prabhavathy, 2022 propose an automated sarcasm detection and classification tool using hyperparameter tuned deep learning (ASDC-HPTDL) model for social media. The proposed ASDC-HPTDL technique primarily involves pre-processing stage to transform the data into useful format. In the next stage of pre-processing, the pre-processed data are converted into the feature vector by Glove Embedding's technique. Then attention bidirectional gated recurrent unit (ABiGRU) technique is utilized to detect and classify sarcasm. In order to boost the detection outcomes of the ABiGRU technique, a hyperparameter tuning process using improved artificial flora algorithm (IAFO) is employed, showing the novelty of the work. The proposed model is validated using the benchmark dataset and the results are examined in terms of precision, recall, accuracy, and F1-score. Nyangaresi et al., 2022 deploy an ANN for target cell selection in a 5G network, and authentication of communicating entities takes place before being admitted in the new cell. It has a low packet loss ratio and latency variations. This scheme is resilient against conventional 5G attack vectors at relatively low costs. Finding drug-target interactions (DTI) is crucial in making new drugs. As new drugs are discovered, there is a significant focus on repurposing existing drugs and incorporating approved drugs. Most computer models for predicting drug-target interactions emphasize binary classification, but the goal is to determine if two drug targets interact. Sharma & Deswal, 2022 compare drug and protein-encoding techniques to predict DT binding affinities. The validation results on the standard dataset show that the proposed model can help with drug discovery by predicting how well DT binds. Finally, Maray et al., 2022; Marzouk et al., 2022 introduce Intelligent Metaheuristics-based Feature Selection model with Optimal ML approach for Malware Detection (IMFSOML-MD) on IoT-enabled MTS. Primarily, IMFSOML-MD technique involves the design of Quantum Invasive Weed Optimization Algorithm-based Feature Selection technique to optimally choose a subset of features. Moreover, an Optimal Wavelet Neural Network (OWNN) model is employed to perform the classification process. The initial parameters of the WNN model are optimally tuned with the help of Colliding Bodies Optimization algorithm, thereby improving the detection performance. The proposed IMFSOML-MD technique was experimentally validated using publicly available CICMalDroid2020 dataset. The results from extensive comparative analysis demonstrated the superiority of the proposed IMFSOML-MD technique over other compared methods in terms of detection performance with maximum accuracy of 98.96%. We want to express our sincere thanks to the editor-in-chief for allowing us to organize this particular issue. The editorial office staffs are excellent, and we thank them for their support. We are also thankful to all the authors who made this special issue possible, and to the reviewers for their thoughtful contributions. Deepak Gupta received a B. Tech. Degree in 2006 from the Guru Gobind Singh Indraprastha University, Delhi, India. He received an M.E. degree in 2010 from Delhi Technological University, India, and PhD degree in 2017 from Dr. APJ Abdul Kalam Technical University (AKTU), Lucknow, India. He completed his Post-Doc from the National Institute of Telecommunications (Inatel), Brazil, in 2018. He has co-authored more than 207 journal articles, including 168 SCI papers and 45 conference articles. He has authored/edited 60 books, published by IEEE-Wiley, Elsevier, Springer, Wiley, CRC Press, DeGruyter, and Katsons. He has filled four Indian patents. He is the convener of the ICICC, ICDAM, ICCCN, ICIIP & DoSCI Springer conferences series. He is Associate Editor of Computer & Electrical Engineering, Expert Systems, Alexandria Engineering Journal, Intelligent Decision Technologies. He is the recipient of the 2021 IEEE System Council Best Paper Award. He has been featured in the list of top 2% scientist/researcher databases worldwide. In India, Rank 1 as a researcher in the field of healthcare applications (as per Google Scholar citation) and Ranked #78 in India among Top Scientists 2022 by Research.com. He is also working towards promoting Startups and also serving as a Startup Consultant. He is also a series editor of “Elsevier Biomedical Engineering” at Academic Press, Elsevier, “Intelligent Biomedical Data Analysis” at De Gruyter, Germany, and “Explainable AI (XAI) for Engineering Applications” at CRC Press. He is appointed as Consulting Editor at Elsevier. Accomplished productive collaborative research with grants of approximately $144000 from various international funding agencies, and he is Co-PI in an International Indo-Russian Joint project of Rs. 1.31 CR from the Department of Science and Technology.

An efficient early detection of diabetic retinopathy using dwarf mongoose optimization based deep belief network

SummaryIn general, diabetic retinopathy (DR) is a common ocular disease that causes damage to the retina due to blood leakage from the vessels. Earlier detection of DR becomes a complicated task and it is necessary to prevent complete blindness. Various physical examinations are employed in DR detection but manual diagnosis results in misclassification results. Therefore, this article proposes a novel technique to predict and classify the DR disease effectively. The significant objective of the proposed approach involves the effective classification of fundus retinal images into two namely, normal (absence of DR) and abnormal (presence of DR). The proposed DR detection utilizes three vital phases namely, the data preprocessing, image augmentation, feature extraction, and classification. Initially, the image preprocessing is done to remove unwanted noises and to enhance images. Then, the preprocessed image is augmented to enhance the size and quality of the training images. This article proposes a novel modified Gaussian convolutional deep belief network based dwarf mongoose optimization algorithm for effective extraction and classification of retinal images. In this article, an ODIR‐2019 dataset is employed in detecting and classifying DR disease. Finally, the experimentation is carried out and the proposed approach achieved 97% of accuracy. This implies that our proposed approach effectively classifies the fundus retinal images.

An Improved Phase Space Reconstruction Method-Based Hybrid Model for Chaotic Traffic Flow Prediction

Traffic flow is chaotic due to nonstationary realistic factors, and revealing the internal nonlinear dynamics of chaotic data and making high-accuracy predictions is the key to traffic control and inducement. Given that high-quality phase space reconstruction is the foundation of predictive modeling. Firstly, an improved C-C method based on the fused norm search domain is proposed to address the issue that the C-C method in the phase space reconstruction algorithm does not meet the Euclidean metric accuracy and reduces the reconstruction quality when the infinite norm metric is used. Secondly, to address the problem of insufficient learning ability of traditional convolutional combinatorial modeling for complex phase space laws of chaotic traffic flow, the high-dimensional phase space features are extracted using the layer-by-layer pretraining mechanism of convolutional deep belief networks (CDBNs), and the temporal features are extracted by combining with long short-term memory (LSTM). Finally, an improved probabilistic dynamic reproduction-based genetic algorithm (PDRGA) is proposed to address the problem of the hybrid model falling into a local optimum when learning the phase space law. Experiments are conducted in three aspects: phase space reconstruction quality analysis, comparison of optimization algorithm convergence, and prediction model performance comparison. The experimentation with two data sets demonstrates that the improved C-C method combines the advantages of the high accuracy metric of the L2 norm with the low operational complexity of the infinite norm, achieving a balance between reconstruction quality and algorithm efficiency. The proposed PDRGA optimization algorithm is a lightweight improvement of the traditional genetic algorithm (GA) and solves the problem that the model tends to fall into a local optimum by optimizing the initial weights of CDBN. Meanwhile, the five error evaluation indexes of the proposed PDRGA-CDBN-LSTM hybrid model are lower than those of the baseline model, providing a new modeling idea for chaotic traffic flow prediction.

Research Based on the Application and Exploration of Artificial Intelligence in the Field of Traditional Music

Traditional Chinese music has undergone trials and tribulations. To date, traditional music has been gradually improved, preserved, and passed down, both in terms of theoretical works and traditional music varieties. However, the current state of traditional music is still a cause for concern. Whether it is scholars engaged in the study of traditional music, universities, or local government agencies, there are still areas that need to be improved individually. The need to cultivate a future audience for traditional music and to make full use of new media such as the Internet is a priority at this stage. The emergence of any new technology is bound to have some impact on the existing social system, and the emergence of artificial intelligence is no exception. Due to the limitation of technology, only a small number of basic applications have been developed, and it is the future mission of research workers whether they can develop more advanced products for music lovers to experience on the basis of ensuring the basic maturity of these applications. In this paper, we introduce the convolutional deep belief network model based on the restricted Boltzmann machine and apply the convolutional deep belief network algorithm to the music melody recognition. Firstly, it was pretrained by an unsupervised greedy layer-by-layer algorithm. Then, the network parameters were fine-tuned by a supervised network training method, and the recognition ability of the model was improved by adjusting the network parameters. The experimental results show that the recognition effect of the system is more obvious under the condition that the length of music samples is 3 s, and the recognition effect is better when the number of model layers is 2 than that when the number of layers is 1.

Effective hybrid deep learning model for COVID-19 patterns identification using CT images.

Coronavirus disease 2019 (COVID‐19) has attracted significant attention of researchers from various disciplines since the end of 2019. Although the global epidemic situation is stabilizing due to vaccination, new COVID‐19 cases are constantly being discovered around the world. As a result, lung computed tomography (CT) examination, an aggregated identification technique, has been used to ameliorate diagnosis. It helps reveal missed diagnoses due to the ambiguity of nucleic acid polymerase chain reaction. Therefore, this study investigated how quickly and accurately hybrid deep learning (DL) methods can identify infected individuals with COVID‐19 on the basis of their lung CT images. In addition, this study proposed a developed system to create a reliable COVID‐19 prediction network using various layers starting with the segmentation of the lung CT scan image and ending with disease prediction. The initial step of the system starts with a proposed technique for lung segmentation that relies on a no‐threshold histogram‐based image segmentation method. Afterward, the GrabCut method was used as a post‐segmentation method to enhance segmentation outcomes and avoid over‐and under‐segmentation problems. Then, three pre‐trained models of standard DL methods, including Visual Geometry Group Network, convolutional deep belief network, and high‐resolution network, were utilized to extract the most affective features from the segmented images that can help to identify COVID‐19. These three described pre‐trained models were combined as a new mechanism to increase the system's overall prediction capabilities. A publicly available dataset, namely, COVID‐19 CT, was used to test the performance of the proposed model, which obtained a 95% accuracy rate. On the basis of comparison, the proposed model outperformed several state‐of‐the‐art studies. Because of its effectiveness in accurately screening COVID‐19 CT images, the developed model will potentially be valuable as an additional diagnostic tool for leading clinical professionals.

A single-phase-to-ground fault location method based on convolutional deep belief network

After single-phase-to-ground fault (SPGF) occurs in the resonant grounding system, it will produce transient electrical quantities with abundant characteristics, such as transient zero-sequence current, transient three-phase current, and transient zero-sequence voltage. At this point, there is a great discrepancy between transient zero-sequence current and transient three-phase current sudden-change at upstream and downstream detection nodes, while those at detection nodes on the same side are similar. Besides, the trend of zero-sequence current at upstream detection nodes and zero-sequence voltage is opposite, while the trend of them downstream is similar. On this basis, this paper presents a new method for locating SPGF, which is based on the convolutional deep belief network (CDBN). Firstly, transient zero-sequence voltage, transient zero-sequence current, and transient three-phase current sudden-change are decomposed and reconstructed by discrete wavelet packet transform (DWPT) to obtain a time-frequency matrix. Then, taking it as input of CDBN to autonomously extract fault characteristics, and fault detection nodes are divided into upstream and downstream detection nodes. Finally, combined with network topology, fault location can be realized. The simulation test and data analysis show this method can identify fault segments accurately and reliably, and has good adaptability and application value.

Deep learning aided beam vector assignment for massive MIMO maritime communication considering location information and handover impact

Location aided beamforming (LAB) has been proposed as a potential solution to fast beam vector (BV) selection for massive multi-input multi-output (M-MIMO) maritime communication systems. However, the existing LAB schemes hardly consider the impact from handover. In this paper, we propose to apply the Gaussian anomaly detection (GAD) to improve the system performance, particularly in the handover zone, by exploiting the information of sea lane and locations of the user equipment (UE). Furthermore, inspired by the rationale of deep learning (DL) aided image processing methods, we first transform the BV assignment problem to a location-based artificial gray value classification problem, and then utilize a convolutional deep belief networks (CDBN) to predict the BV indices from a predefined codebook constituted by location-related feature matrices. In addition, we exploit GAD to rectify the training data of CDBN, such that the prediction on BV assignments can be more accurate to achieve a higher capacity than known methods. Finally, an extensive comparison with selected existing schemes through simulations demonstrates the effectiveness of the proposed scheme.