Deep Convolutional Neural Network Approach Research Articles

Spoken emotion recognition through human-computer interaction using a novel deep learning technology

The paradigm of textual or display-based control in human-computer interaction (HCI) has changed in favor of more understandable control methods, such as gesture, voice, and imitation. Speech in particular contains a large quantity of information, revealing the speaker's inner state as well as his or her goal and intention. The speaker's request can be understood through language analysis, but additional speech features show the speaker's mood, purpose, and intention. As a consequence, in modern HCI systems, emotion identification from speech has become crucial. Additionally, it is challenging to aggregate the results of the many professionals engaged in emotion identification. There have been several methods for analyzing sound in the past. However, it was impossible to analyses people's emotions during a live speech. Studies on real-time data are now more prominent than ever because of the advancement of artificial intelligence and the great performance of deep learning techniques. This research uses a cutting-edge deep-learning technique to identify emotions in human speech. The research made use of the open-source Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) dataset. More than 2000 fragments of data were captured by 24 performers as speeches and songs for the RAVDESS dataset. The actors' responses to eight distinct moods were recorded. It was designed to find various emotion classifications. In this study, a novel neuro-fuzzy swallow swarm-optimized deep convolutional neural networks (NFSO-DCNN) approach for classification was suggested. The performance of the suggested model was compared to that of similar research, and the outcomes were assessed. Employing the suggested example on the RAVDESS dataset, an overall accuracy of 98.5% was attained for categorizing emotions

Combining ERA5 data and CYGNSS observations for the joint retrieval of global significant wave height of ocean swell and wind wave: a deep convolutional neural network approach

Combining ERA5 data and CYGNSS observations for the joint retrieval of global significant wave height of ocean swell and wind wave: a deep convolutional neural network approach

An effective human monkeypox classification using vision transformer

AbstractMonkeypox virus has become a popular disease as many people have been recently infected with this virus around the world. This virus causes fears of another global pandemic similar to COVID‐19 pandemic. Therefore, it is important to determine infected people at an early stage to reduce the spread of this virus. Creating an automatic approach to detect monkeypox virus is a challenging problem due to the similar appearance between skin infections. Currently, deep convolutional neural network approaches have been used in detecting and classifying skin infections. However, these approaches sever from certain challenges such as being unable to capture the global context and low accuracy rate. In this paper, we propose a novel approach based on vision transformer to automatically detect monkeypox virus in skin images. We fine‐tune vision transformer for the human monkeypox classification problems. This fine‐tuned vision transformer splits an image into image patches and then these patches are used as input to the transformer in a sequence structure, followed by added layers to detect monkeypox virus. The experimental results show that the proposed approach achieves excellent detection accuracy using a public dataset.

Deep Learning-assisted Retinopathy of Prematurity (ROP) Screening

Retinopathy of prematurity (ROP) is a leading cause of blindness in premature infants worldwide, particularly in developing countries. In this research, we propose a Deep Convolutional Neural Network (DCNN) and image processing-based approach for the automatic detection of retinal features, including the optical disc (OD) and retinal blood vessels (BV), as well as disease classification using a rule-based method for ROP patients. Our DCNN model uses YOLO-v5 for OD detection and either Pix2Pix or a U-Net for BV segmentation. We trained our DCNN models on publicly available fundus image datasets of size 1,117 and 288 for OD detection and BV segmentation, respectively. We evaluated our approach on a dataset of 439 preterm neonatal retinal images, testing for ROP Zone and 6 BV masks. Our proposed system achieved excellent results, with the OD detection module achieving an overall accuracy of 98.94% (when IoU 0.5) and the BV segmentation module achieving an accuracy of 96.69% and a Dice coefficient between 0.60 and 0.64. Moreover, our system accurately diagnosed ROP in Zone-1 with 88.23% accuracy. Our approach offers a promising solution for accurate ROP screening and diagnosis, particularly in low-resource settings, where it has the potential to improve healthcare outcomes.

A Leading but Simple Classification Method for Remote Sensing Images

Recently, researchers have proposed a lot of deep convolutional neural network (CNN) approaches with obvious flaws to tackle the difficult semantic classification (SC) task of remote sensing images (RSI). In this paper, the author proposes a simple method that aims to provide a leading but efficient solution by using a lightweight EfficientNet-B0. First, this paper concluded the drawbacks with an analysis of mathematical theory and then proposed a qualitative conclusion on the previous methods’ theoretical performance based on theoretical derivation and experiments. Following that, the paper designs a novel method named LS-EfficientNet, consisting only of a single CNN and a concise training algorithm called SC-CNN. Far different from previous complex and hardware-extensive ones, the proposed method mainly focuses on tackling the long-neglected problems, including overfitting, data distribution shift by DA, improper use of training tricks, and other incorrect operations on a pre-trained CNN. Compared to previous studies, the proposed method is easy to reproduce because all the models, training tricks, and hyperparameter settings are open-sourced. Extensive experiments on two benchmark datasets show that the proposed method can easily surpass all the previous state-of-the-art ones, with an outstanding accuracy lead of 0.5% to 1.2% and a remarkable parameter decrease of 78% if compared to the best prior one in 2022. In addition, ablation test results also prove that the proposed effective combination of training tricks, including OLS and CutMix, can clearly boost a CNN's performance for RSI-SC, with an increase in accuracy of 1.0%. All the results reveal that a single lightweight CNN can well tackle the routine task of classifying RSI.

FireClassNet: a deep convolutional neural network approach for PJF fire images classification

FireClassNet: a deep convolutional neural network approach for PJF fire images classification

A transfer learning‐based deep convolutional neural network approach for induction machine multiple faults detection

SummaryThe condition monitoring of squirrel cage induction motors (SCIMs) is vital for uninterrupted production and minimum downtime. Early fault detection can boost output with minimum effort. This article combines the application of transfer learning and convolution neural network (TL‐CNN) for developing an efficient model for bearing and rotor broken bars damage identification in SCIMs. A simple technique for the 1‐D current signal‐to‐image conversion is also proposed to provide input to the proposed deep learning‐based TL‐CNN technique. The proposed approach embodies the advantages of TL and CNN for effective fault identification in SCIMs. The developed technique has classified faults efficiently with an average accuracy of 99.40%. The complete analysis and data collection have been done on the experimental set‐up with a 5 kW SCIM and LabVIEW‐based data acquisition system. The propounded fault detection model has been created in python with the help of packages like Keras and TensorFlow.

Electromagnetic Imaging for Buried Conductors Using Deep Convolutional Neural Networks

In the past, many conventional algorithms, such as self-adaptive dynamic differential evolution and asynchronous particle swarm optimization, were used to reconstruct buried objects in the frequency domain; these were unfortunately time-consuming during the iterative, repeated computing process of the scattered field. Consequently, we propose an innovative deep convolutional neural network approach to solve the electromagnetic inverse scattering problem for buried conductors in this paper. Different shapes of conductors are buried in one half-space and the electromagnetic wave from the other half-space is incident. The shape of the conductor can be reconstructed promptly by inputting the received scattered fields measured from the upper half-space into the deep convolutional neural network module, which avoids the computational complexity of Green’s function for training. Numerical results show that the root mean square error for differently shaped—circular, elliptical, arrow, peanut, four-petal, and three-petal—reconstructed images are, respectively, 2.95%, 3.11%, 17.81%, 15.10%, 14.14%, and 15.24%. Briefly speaking, not only can circular and elliptical buried conductors be reconstructed; some irregular shapes can be reconstructed well. On the contrary, the reconstruction result by U-Net for buried objects is worse since it is not able to obtain a good preliminary image by processing only the upper scattered field—that is, rather than the full space. In other words, our proposed deep convolutional neural network can efficiently solve the electromagnetic inverse scattering problem of buried conductors and provide a novel method for the microwave imaging of the buried conductors. This is the first successful attempt at using deep convolutional neural networks for buried conductors in the frequency domain, which may be useful for practical applications in various fields such as the medical, military, or industrial fields, including magnetic resonance imaging, mine detection and clearance, non-destructive testing, gas or wire pipeline detection, etc.

Mathematical model and artificial intelligence for diagnosis of Alzheimer's disease.

Degeneration of the neurological system linked to cognitive deficits, daily living exercise clutters, and behavioral disturbing impacts may define Alzheimer's disease. Alzheimer's disease research conducted later in life focuses on describing ways for early detection of dementia, a kind of mental disorder. To tailor our care to each patient, we utilized visual cues to determine how they were feeling. We did this by outlining two approaches to diagnosing a person's mental health. Support vector machine is the first technique. Image characteristics are extracted using a fractal model for classification in this method. With this technique, the histogram of a picture is modeled after a Gaussian distribution. Classification was performed with several support vector machines kernels, and the outcomes were compared. Step two proposes using a deep convolutional neural network architecture to identify Alzheimer's-related mental disorders. According to the findings, the support vector machines approach accurately recognized over 93% of the photos tested. The deep convolutional neural network approach was one hundred percent accurate during model training, whereas the support vector machines approach achieved just 93 percent accuracy. In contrast to support vector machines accuracy of 89.3%, the deep convolutional neural network model test findings were accurate 98.8% of the time. Based on the findings reported here, the proposed deep convolutional neural network architecture may be used for diagnostic purposes involving the patient's mental state.

Design of a novel masking-based deep convolutional neural network approach for accident classification

This work presents a novel multi-class detection approach for predicting car damage due to accidents or other reasons. The proposed prediction model intends to extract the class information to analyze the objects over the frames. Here, a novel masking-based deep convolutional neural network (MD-CNN) is proposed to capture the car parts’ regions and the classification process. The proposed model works well compared to the various existing approaches. The computation is initiated with the acquisition of damaged images, and pre-processing is performed for data labeling, which is further divided into training and testing samples. The features are internally analyzed by the proposed MD-CNN model, where the internal feature extraction process is a significant advantage with the classification process. The simulation is done in MATLAB 2020a environment where various performance metrics like accuracy, precision, [Formula: see text]-measure, recall and detection rate are evaluated and compared with conventional learning approaches. The model gives an average precision of 94.5% and [Formula: see text]1-score of 91% which is higher than other approaches. The proposed model establishes a better trade-off among the prevailing approaches with superior prediction accuracy.

Investigation of radiomics and deep convolutional neural networks approaches for glioma grading

Purpose. To determine glioma grading by applying radiomic analysis or deep convolutional neural networks (DCNN) and to benchmark both approaches on broader validation sets. Methods. Seven public datasets were considered: (1) low-grade glioma or high-grade glioma (369 patients, BraTS’20) (2) well-differentiated liposarcoma or lipoma (115, LIPO); (3) desmoid-type fibromatosis or extremity soft-tissue sarcomas (203, Desmoid); (4) primary solid liver tumors, either malignant or benign (186, LIVER); (5) gastrointestinal stromal tumors (GISTs) or intra-abdominal gastrointestinal tumors radiologically resembling GISTs (246, GIST); (6) colorectal liver metastases (77, CRLM); and (7) lung metastases of metastatic melanoma (103, Melanoma). Radiomic analysis was performed on 464 (2016) radiomic features for the BraTS’20 (others) datasets respectively. Random forests (RF), Extreme Gradient Boosting (XGBOOST) and a voting algorithm comprising both classifiers were tested. The parameters of the classifiers were optimized using a repeated nested stratified cross-validation process. The feature importance of each classifier was computed using the Gini index or permutation feature importance. DCNN was performed on 2D axial and sagittal slices encompassing the tumor. A balanced database was created, when necessary, using smart slices selection. ResNet50, Xception, EficientNetB0, and EfficientNetB3 were transferred from the ImageNet application to the tumor classification and were fine-tuned. Five-fold stratified cross-validation was performed to evaluate the models. The classification performance of the models was measured using multiple indices including area under the receiver operating characteristic curve (AUC). Results. The best radiomic approach was based on XGBOOST for all datasets; AUC was 0.934 (BraTS’20), 0.86 (LIPO), 0.73 (LIVER), (0.844) Desmoid, 0.76 (GIST), 0.664 (CRLM), and 0.577 (Melanoma) respectively. The best DCNN was based on EfficientNetB0; AUC was 0.99 (BraTS’20), 0.982 (LIPO), 0.977 (LIVER), (0.961) Desmoid, 0.926 (GIST), 0.901 (CRLM), and 0.89 (Melanoma) respectively. Conclusion. Tumor classification can be accurately determined by adapting state-of-the-art machine learning algorithms to the medical context.

Whale- crow search optimisation enabled deep convolutional neural network for flood detection

ABSTRACT The satellite images are more attracted in the field of flood detection. For planning actions during emergencies, flood detection plays a vital role, but the major barrier is that using satellite images to detect flooded regions. For flood detection, this method innovates a model named Whale-crow search algorithm on the basis of deep convolutional neural network (W-CSA DCNN) approach. Pre-processing, classification, segmentation and feature extraction are the four steps which is included in this model. For obtaining sound and antiquity from the input image initially, the satellite imagery is given to pre-processing and then for obtaining the features on the basis of vegetation indices the pre-processed image is put through the feature extraction process. By means of Kernel Fuzzy Auto regressive (KFAR) model, the acquire features are subsequently used in the segmentation process. After obtaining the segments, it is given to the classification, which is carried out by means of DCNN and qualified excellently via the W-CSA that is the combination of the Crow Search Algorithm (CSA) and Whale optimisation algorithm (WOA). Based on the specificity, accuracy and sensitivity with values 0.982, 0.972 and 0.975, the efficiency of this process deliberates advanced performance than the existing process.

TSMC-Net: Deep-Learning Multigas Classification Using THz Absorption Spectra.

The identification of gas mixture speciation from a complex multicomponent absorption spectrum is a problem in gas sensing that can be addressed using machine-learning approaches. Here, we report on a deep convolutional neural network for multigas classification using terahertz (THz) absorption spectra, THz spectra mixture classifier network or TSMC-Net. TSMC-Net has been developed to identify eight volatile organic compounds in mixtures based on their fingerprint rotational absorption spectra in the 220-330 GHz frequency range. A data set consisting of simulated absorption spectra for randomly generated mixtures, with absorption greater than thresholds representing detectable limits and annotated with multiple labels, was prepared for model development. The supervised multilabel classification problem, i.e., the identification of individual gases in a mixture, is converted to a supervised multiclass classification problem via label powerset conversion. The trained model is validated and tested against simulated spectra for gas mixtures, with and without white Gaussian noise. The trained model exhibits high precision, recall, and accuracy for each pure compound. Class activation maps illustrate the complex decision-making process of the model and highlight relevant frequency regions that are needed to identify unique mixtures. Finally, the model was demonstrated against measured THz absorption spectra for pure species and mixtures, acquired using a microelectronics-based THz absorption spectrometer. The data set generation strategy and deep convolutional neural network approach are generalized and can be extrapolated to other spectroscopy types, frequency ranges, and sensors.

Mineral Texture Classification Using Deep Convolutional Neural Networks: An Application to Zircons From Porphyry Copper Deposits

AbstractThe texture and morphology of igneous zircon indicates magmatic conditions during zircon crystallization and can be used to constrain provenance. Zircons from porphyry copper deposits are typically prismatic, euhedral, and strongly oscillatory zoned which may differentiate them from zircons associated with unmineralized igneous systems. Here, cathodoluminescence images of zircons from the Quellaveco porphyry copper district, Southern Peru, were collected to compare zircon textures between the premineralization Yarabamba Batholith and the Quellaveco porphyry copper deposit. Quellaveco porphyry zircons are prismatic, euhedral, and strongly oscillatory zoned, whereas the batholith zircons are subhedral‐anhedral with weaker zoning. We adopt a deep convolutional neural network (CNN) approach to demonstrate that a CNN can classify Quellaveco porphyry zircons with high success. We trial several CNN architectures to classify zircon images: LeNet‐5, AlexNet and VGG, including a transfer learning approach where we used the weights of a VGG model pretrained on the ImageNet data set. The VGG model with transfer learning is the most effective approach, with accuracy and receiver operating characteristic‐area under curve (ROC‐AUC) scores of 0.86 and 0.93, indicating that a Quellaveco porphyry zircon CL image can be ranked higher than a batholith zircon with 93% probability. Visualizing model layer outputs demonstrates that the CNN models can recognize crystal edges, zoning, and mineral inclusions. We trial implementing trained CNN models as unsupervised feature extractors, which can empirically quantify crystal textures and morphology. Therefore, deep learning provides a tool for the extraction of information from large, imaged‐based petrographic data sets which can facilitate petrologic and provenance studies.

Prior knowledge auxiliary for few-shot pest detection in the wild.

One of the main techniques in smart plant protection is pest detection using deep learning technology, which is convenient, cost-effective, and responsive. However, existing deep-learning-based methods can detect only over a dozen common types of bulk agricultural pests in structured environments. Also, such methods generally require large-scale well-labeled pest data sets for their base-class training and novel-class fine-tuning, and these significantly hinder the further promotion of deep convolutional neural network approaches in pest detection for economic crops, forestry, and emergent invasive pests. In this paper, a few-shot pest detection network is introduced to detect rarely collected pest species in natural scenarios. Firstly, a prior-knowledge auxiliary architecture for few-shot pest detection in the wild is presented. Secondly, a hierarchical few-shot pest detection data set has been built in the wild in China over the past few years. Thirdly, a pest ontology relation module is proposed to combine insect taxonomy and inter-image similarity information. Several experiments are presented according to a standard few-shot detection protocol, and the presented model achieves comparable performance to several representative few-shot detection algorithms in terms of both mean average precision (mAP) and mean average recall (mAR). The results show the promising effectiveness of the proposed few-shot detection architecture.

Machine learning based small bowel video capsule endoscopy analysis: Challenges and opportunities

Video capsule endoscopy (VCE) is a revolutionary technology for the early diagnosis of gastric disorders. However, owing to the high redundancy and subtle manifestation of anomalies among thousands of frames, the manual construal of VCE videos requires considerable patience, focus, and time. The automatic analysis of these videos using computational methods is a challenge as the capsule is untamed in motion and captures frames inaptly. Several machine learning (ML) methods, including recent deep convolutional neural networks approaches, have been adopted after evaluating their potential of improving the VCE analysis. However, the clinical impact of these methods is yet to be investigated. This survey aimed to highlight the gaps between existing ML-based research methodologies and clinically significant rules recently established by gastroenterologists based on VCE. A framework for interpreting raw frames into contextually relevant frame-level findings and subsequently merging these findings with meta-data to obtain a disease-level diagnosis was formulated. Frame-level findings can be more intelligible for discriminative learning when organized in a taxonomical hierarchy. The proposed taxonomical hierarchy, which is formulated based on pathological and visual similarities, may yield better classification metrics by setting inference classes at a higher level than training classes. Mapping from the frame level to the disease level was structured in the form of a graph based on clinical relevance inspired by the recent international consensus developed by domain experts. Furthermore, existing methods for VCE summarization, classification, segmentation, detection, and localization were critically evaluated and compared based on aspects deemed significant by clinicians. Numerous studies pertain to single anomaly detection instead of a pragmatic approach in a clinical setting. The challenges and opportunities associated with VCE analysis were delineated. A focus on maximizing the discriminative power of features corresponding to various subtle lesions and anomalies may help cope with the diverse and mimicking nature of different VCE frames. Large multicenter datasets must be created to cope with data sparsity, bias, and class imbalance. Explainability, reliability, traceability, and transparency are important for an ML-based diagnostics system in a VCE. Existing ethical and legal bindings narrow the scope of possibilities where ML can potentially be leveraged in healthcare. Despite these limitations, ML based video capsule endoscopy will revolutionize clinical practice, aiding clinicians in rapid and accurate diagnosis.

IRMIRS: Inception-ResNet-Based Network for MRI Image Super-Resolution

Medical image super-resolution is a fundamental challenge due to absorption and scattering in tissues. These challenges are increasing the interest in the quality of medical images. Recent research has proven that the rapid progress in convolutional neural networks (CNNs) has achieved superior performance in the area of medical image super-resolution. However, the traditional CNN approaches use interpolation techniques as a preprocessing stage to enlarge low-resolution magnetic resonance (MR) images, adding extra noise in the models and more memory consumption. Furthermore, conventional deep CNN approaches used layers in series-wise connection to create the deeper mode, because this later end layer cannot receive complete information and work as a dead layer. In this paper, we propose Inception-ResNet-based Network for MRI Image Super-Resolution known as IRMRIS. In our proposed approach, a bicubic interpolation is replaced with a deconvolution layer to learn the upsampling filters. Furthermore, a residual skip connection with the Inception block is used to reconstruct a high-resolution output image from a low-quality input image. Quantitative and qualitative evaluations of the proposed method are supported through extensive experiments in reconstructing sharper and clean texture details as compared to the state-of-the-art methods.

Histogram Matched Chest X-Rays Based Tuberculosis Detection Using CNN

Tuberculosis (TB) is a severe infection that mostly affects the lungs and kills millions of people’s lives every year. Tuberculosis can be diagnosed using chest X-rays (CXR) and data-driven deep learning (DL) approaches. Because of its better automated feature extraction capability, convolutional neural networks (CNNs) trained on natural images are particularly effective in image categorization. A combination of 3001 normal and 3001 TB CXR images was gathered for this study from different accessible public datasets. Ten different deep CNNs (Resnet50, Resnet101, Resnet152, InceptionV3, VGG16, VGG19, DenseNet121, DenseNet169, DenseNet201, MobileNet) are trained and tested for identifying TB and normal cases. This study presents a deep CNN approach based on histogram matched CXR images that does not require object segmentation of interest, and this coupled methodology of histogram matching with the CXRs improves the accuracy and detection performance of CNN models for TB detection. Furthermore, this research contains two separate experiments that used CXR images with and without histogram matching to classify TB and non-TB CXRs using deep CNNs. It was able to accurately detect TB from CXR images using pre-processing, data augmentation, and deep CNN models. Without histogram matching the best accuracy, sensitivity, specificity, precision and F1-score in the detection of TB using CXR images among ten models are 99.25%, 99.48%, 99.52%, 99.48% and 99.22% respectively. With histogram matching the best accuracy, sensitivity, specificity, precision and F1-score are 99.58%, 99.82%, 99.67%, 99.65% and 99.56% respectively. The proposed methodology, which has cutting-edge performance, will be useful in computer-assisted TB diagnosis and aids in minimizing irregularities in TB detection in developing countries.

Deep Convolutional Neural Network for Detection and Prediction of Waxy Corn Seed Viability Using Hyperspectral Reflectance Imaging

This paper aimed to combine hyperspectral imaging (378–1042 nm) and a deep convolutional neural network (DCNN) to rapidly and non-destructively detect and predict the viability of waxy corn seeds. Different viability levels were set by artificial aging (aging: 0 d, 3 d, 6 d, and 9 d), and spectral data for the first 10 h of seed germination were continuously collected. Bands that were significantly correlated (SC) with moisture, protein, starch, and fat content in the seeds were selected, and another optimal combination was extracted using a successive projection algorithm (SPA). The support vector machine (SVM), k-nearest neighbor (KNN), random forest (RF), and deep convolutional neural network (DCNN) approaches were used to establish the viability detection and prediction models. During detection, with the addition of different levels, the recognition effect of the first three methods decreased, while the DCNN method remained relatively stable (always above 95%). When using the previous 2.5 h data, the prediction accuracy rate was generally higher than the detection model. Among them, SVM + full band increased the most, while DCNN + full band was the highest, reaching 98.83% accuracy. These results indicate that the combined use of hyperspectral imaging technology and the DCNN method is more conducive to the rapid detection and prediction of seed viability.

Geographical classification of malaria parasites through applying machine learning to whole genome sequence data

Malaria, caused by Plasmodium parasites, is a major global health challenge. Whole genome sequencing (WGS) of Plasmodium falciparum and Plasmodium vivax genomes is providing insights into parasite genetic diversity, transmission patterns, and can inform decision making for clinical and surveillance purposes. Advances in sequencing technologies are helping to generate timely and big genomic datasets, with the prospect of applying Artificial Intelligence analytical techniques (e.g., machine learning) to support programmatic malaria control and elimination. Here, we assess the potential of applying deep learning convolutional neural network approaches to predict the geographic origin of infections (continents, countries, GPS locations) using WGS data of P. falciparum (n = 5957; 27 countries) and P. vivax (n = 659; 13 countries) isolates. Using identified high-quality genome-wide single nucleotide polymorphisms (SNPs) (P. falciparum: 750 k, P. vivax: 588 k), an analysis of population structure and ancestry revealed clustering at the country-level. When predicting locations for both species, classification (compared to regression) methods had the lowest distance errors, and > 90% accuracy at a country level. Our work demonstrates the utility of machine learning approaches for geo-classification of malaria parasites. With timelier WGS data generation across more malaria-affected regions, the performance of machine learning approaches for geo-classification will improve, thereby supporting disease control activities.