Classification Recognition Research Articles

MES-CTNet: A Novel Capsule Transformer Network Base on a Multi-Domain Feature Map for Electroencephalogram-Based Emotion Recognition.

Emotion recognition using the electroencephalogram (EEG) has garnered significant attention within the realm of human-computer interaction due to the wealth of genuine emotional data stored in EEG signals. However, traditional emotion recognition methods are deficient in mining the connection between multi-domain features and fitting their advantages. In this paper, we propose a novel capsule Transformer network based on a multi-domain feature for EEG-based emotion recognition, referred to as MES-CTNet. The model's core consists of a multichannel capsule neural network(CapsNet) embedded with ECA (Efficient Channel Attention) and SE (Squeeze and Excitation) blocks and a Transformer-based temporal coding layer. Firstly, a multi-domain feature map is constructed by combining the space-frequency-time characteristics of the multi-domain features as inputs to the model. Then, the local emotion features are extracted from the multi-domain feature maps by the improved CapsNet. Finally, the Transformer-based temporal coding layer is utilized to globally perceive the emotion feature information of the continuous time slices to obtain a final emotion state. The paper fully experimented on two standard datasets with different emotion labels, the DEAP and SEED datasets. On the DEAP dataset, MES-CTNet achieved an average accuracy of 98.31% in the valence dimension and 98.28% in the arousal dimension; it achieved 94.91% for the cross-session task on the SEED dataset, demonstrating superior performance compared to traditional EEG emotion recognition methods. The MES-CTNet method, utilizing a multi-domain feature map as proposed herein, offers a broader observation perspective for EEG-based emotion recognition. It significantly enhances the classification recognition rate, thereby holding considerable theoretical and practical value in the EEG emotion recognition domain.

Unlocking the Future of Secure Automatic Machines: Leveraging FaceReg with HRC & LBPH

We propose a Computer vision and Machine Learning equipped model which secures ATMs from fraudulent activities by leveraging the use of Haar cascade (HRC) and Local Binary Pattern Histogram (LBPH) classifier for face detection and recognition correspondingly, which in turn detect fraud by utilizing features like PIN and also face recognition helps to identify and authenticate the user by checking with the trained dataset and trigger real-time alert mail if the user finds to be unauthorized also, do not allow them to log in into the machine which resolves the ATM security issue. this system is evaluated on the dataset of real-world ATM camera feeds, which shows an accuracy of 90%. It can effectively detect many frauds, including identity theft and unauthorized access which makes it even more reliable.

Optical coherence tomography for multicellular tumor spheroid category recognition and drug screening classification via multi-spatial-superficial-parameter and machine learning.

Optical coherence tomography (OCT) is an ideal imaging technique for noninvasive and longitudinal monitoring of multicellular tumor spheroids (MCTS). However, the internal structure features within MCTS from OCT images are still not fully utilized. In this study, we developed cross-statistical, cross-screening, and composite-hyperparameter feature processing methods in conjunction with 12 machine learning models to assess changes within the MCTS internal structure. Our results indicated that the effective features combined with supervised learning models successfully classify OVCAR-8 MCTS culturing with 5,000 and 50,000 cell numbers, MCTS with pancreatic tumor cells (Panc02-H7) culturing with the ratio of 0%, 33%, 50%, and 67% of fibroblasts, and OVCAR-4 MCTS treated by 2-methoxyestradiol, AZD1208, and R-ketorolac with concentrations of 1, 10, and 25 µM. This approach holds promise for obtaining multi-dimensional physiological and functional evaluations for using OCT and MCTS in anticancer studies.

A Comprehensive Framework for Long-Tailed Learning via Pretraining and Normalization.

Data in the visual world often present long-tailed distributions. However, learning high-quality representations and classifiers for imbalanced data is still challenging for data-driven deep learning models. In this work, we aim at improving the feature extractor and classifier for long-tailed recognition via contrastive pretraining and feature normalization, respectively. First, we carefully study the influence of contrastive pretraining under different conditions, showing that current self-supervised pretraining for long-tailed learning is still suboptimal in both performance and speed. We thus propose a new balanced contrastive loss and a fast contrastive initialization scheme to improve previous long-tailed pretraining. Second, based on the motivative analysis on the normalization for classifier, we propose a novel generalized normalization classifier that consists of generalized normalization and grouped learnable scaling. It outperforms traditional inner product classifier as well as cosine classifier. Both the two components proposed can improve recognition ability on tail classes without the expense of head classes. We finally build a unified framework that achieves competitive performance compared with state of the arts on several long-tailed recognition benchmarks and maintains high efficiency.

An Aero-Engine Classification Method Based on Fourier Transform Infrared Spectrometer Spectral Feature Vectors

Aiming at the classification identification problem of aero-engines, this paper adopts a telemetry Fourier transform infrared spectrometer to collect aero-engine hot jet infrared spectrum data and proposes an aero-engine classification identification method based on spectral feature vectors. First, aero-engine hot jet infrared spectrum data are acquired and measured; meanwhile, the spectral feature vectors based on CO2 are constructed. Subsequently, the feature vectors are combined with the seven mainstream classification algorithms to complete the training and prediction of the classification model. In the experiment, two Fourier transform infrared spectrometers, EM27 developed by Bruker and a self-developed telemetry FT-IR spectrometer, were used to telemeter the hot jet of three aero-engines to obtain infrared spectral data. The training data set and test data set were randomly divided in a ratio of 3:1. The model training of the training data set and the label prediction of the test data set were carried out by combining spectral feature vectors and classification algorithms. The classification evaluation indicators were accuracy, precision, recall, confusion matrix, and F1-score. The classification recognition accuracy of the algorithm was 98%. This paper has considerable significance for the fault diagnosis of aero-engines and classification recognition of aircrafts.

TLN-LSTM: an automatic modulation recognition classifier based on a two-layer nested structure of LSTM network for extremely long signal sequences

Purpose Automatic modulation recognition (AMR) is a challenging problem in intelligent communication systems and has wide application prospects. At present, although many AMR methods based on deep learning have been proposed, the methods proposed by these works cannot be directly applied to the actual wireless communication scenario, because there are usually two kinds of dilemmas when recognizing the real modulated signal, namely, long sequence and noise. This paper aims to effectively process in-phase quadrature (IQ) sequences of very long signals interfered by noise. Design/methodology/approach This paper proposes a general model for a modulation classifier based on a two-layer nested structure of long short-term memory (LSTM) networks, called a two-layer nested structure (TLN)-LSTM, which exploits the time sensitivity of LSTM and the ability of the nested network structure to extract more features, and can achieve effective processing of ultra-long signal IQ sequences collected from real wireless communication scenarios that are interfered by noise. Findings Experimental results show that our proposed model has higher recognition accuracy for five types of modulation signals, including amplitude modulation, frequency modulation, gaussian minimum shift keying, quadrature phase shift keying and differential quadrature phase shift keying, collected from real wireless communication scenarios. The overall classification accuracy of the proposed model for these signals can reach 73.11%, compared with 40.84% for the baseline model. Moreover, this model can also achieve high classification performance for analog signals with the same modulation method in the public data set HKDD_AMC36. Originality/value At present, although many AMR methods based on deep learning have been proposed, these works are based on the model’s classification results of various modulated signals in the AMR public data set to evaluate the signal recognition performance of the proposed method rather than collecting real modulated signals for identification in actual wireless communication scenarios. The methods proposed in these works cannot be directly applied to actual wireless communication scenarios. Therefore, this paper proposes a new AMR method, dedicated to the effective processing of the collected ultra-long signal IQ sequences that are interfered by noise.

Recognition of Mushrooms and Classification of Edible and Toxic Families using Hardware Implementation of CNN Algorithms on an Embedded system

In this work, new ideas in the realm of picture identification and classification are developed and implemented on hardware. This entails putting new algorithms into practice, whether for color, texture, or shape identification for AI (Artificial Intelligence) and picture recognition applications. We concentrate on identifying edible mushrooms in the harvesting and food manufacturing processes. Our proposal for an embedded system based on a Raspberry-Pi4 type microcomputer employing a combination of hardware and software components has helped with the recognition and classification of items in the image. Our object recognition system is built on a novel neighborhood topology and a cutting-edge kernel function that enables the effective embedding of image processing-related characteristics. We tested the suggested CNN-based object recognition system using a variety of challenging settings, including diverse fungus species, uncontrolled environments, and varying backdrop and illumination conditions. The outcomes were superior to various state-of-the-art outcomes. On the other hand, our contribution relating to the dynamic mode integrates a CNN network to accurately encode the temporal information with an attention mask allowing us to focus on the characteristics of an edible mushroom according to the state of the art, and guarantee the robustness of the recognition. We implemented our algorithm on a Raspberry Pi400-based embedded system connected to a CMOS camera-type image sensor plus an HMI human-machine interface for the instantaneous display of results for the rapid classification of edible and inedible mushrooms.

PNER: Applying the Pipeline Method to Resolve Nested Issues in Named Entity Recognition

Named entity recognition (NER) in natural language processing encompasses three primary types: flat, nested, and discontinuous. While the flat type often garners attention from researchers, nested NER poses a significant challenge. Current approaches to addressing nested NER involve sequence labeling methods with merged label layers, cascaded models, and those rooted in reading comprehension. Among these, sequence labeling with merged label layers stands out for its simplicity and ease of implementation. Yet, highlighted issues persist within this method, prompting our aim to enhance its efficacy. In this study, we propose augmentations to the sequence labeling approach by employing a pipeline model bifurcated into sequence labeling and text classification tasks. Departing from annotating specific entity categories, we amalgamated types into main and sub-categories for a unified treatment. These categories were subsequently embedded as identifiers in the recognition text for the text categorization task. Our choice of resolution involved BERT+BiLSTM+CRF for sequence labeling and the BERT model for text classification. Experiments were conducted across three nested NER datasets: GENIA, CMeEE, and GermEval 2014, featuring annotations varying from four to two levels. Before model training, we conducted separate statistical analyses on nested entities within the medical dataset CMeEE and the everyday life dataset GermEval 2014. Our research unveiled a consistent dominance of a particular entity category within nested entities across both datasets. This observation suggests the potential utility of labeling primary and subsidiary entities for effective category recognition. Model performance was evaluated based on F1 scores, considering correct recognition only when both the complete entity name and category were identified. Results showcased substantial performance enhancement after our proposed modifications compared to the original method. Additionally, our improved model exhibited strong competitiveness against existing models. F1 scores on the GENIA, CMeEE, and GermEval 2014 datasets reached 79.21, 66.71, and 87.81, respectively. Our research highlights that, while preserving the original method’s simplicity and implementation ease, our enhanced model achieves heightened performance and competitive prowess compared to other methodologies.

Clothing Style Similarity Matching and Classification Recognition Algorithm Based on Double-Layer Model in the Context of the Internet of Things

Abstract The identification of different clothing style attributes is helpful for designers to grasp the clothing style, and it is also helpful for consumers to find clothes with the same style attributes according to their own preferences. This is becoming more important in the clothing design, Internet, and e-commerce industries. It is of great significance to carry out clothing style similarity matching and classification recognition. For the problem of clothing style similarity matching and classification recognition, the traditional algorithm stays in the stage of qualitative analysis and subjective evaluation, unable to quantitatively and objectively determine clothing style, resulting in the decline of clothing style similarity matching effect and classification recognition accuracy. Therefore, a similarity matching, classification, and recognition algorithm of clothing style based on the double-layer model in the context of the Internet of Things is designed. The double-layer model is constructed through the target detection layer and target segmentation layer. The double-layer model is used to realize the segmentation of clothing image. The overall similarity evaluation index of image style is obtained according to the image gradient to complete the similarity matching of clothing style. On this basis, the clothing image features are extracted and input into the support vector machine classifier to complete the clothing classification and recognition. The experimental results show that the proposed garment style similarity matching and classification recognition algorithm based on the two-layer model has high image segmentation accuracy, good style similarity matching effect, high classification recognition accuracy, and high efficiency, which proves that the algorithm is feasible for garment style similarity matching and classification recognition using the two-layer model and can be further applied in the field of garment design.

IoT-assisted Human Activity Recognition Using Bat Optimization Algorithm with Ensemble Voting Classifier for Disabled Persons

Internet of Things (IoT)-based human action recognition (HAR) has made a significant contribution to scientific studies. Furthermore, hand gesture recognition is a subsection of HAR, and plays a vital role in interacting with deaf people. It is the automatic detection of the actions of one or many subjects using a series of observations. Convolutional neural network structures are often utilized for finding human activities. With this intention, this study presents a new bat optimization algorithm with an ensemble voting classifier for human activity recognition (BOA-EVCHAR) technique to help disabled persons in the IoT environment. The BOA-EVCHAR technique makes use of the ensemble classification concept to recognize human activities proficiently in the IoT environment. In the presented BOA-EVCHAR approach, data preprocessing is generally achieved at the beginning level. For the identification and classification of human activities, an ensemble of two classifiers namely long short-term memory (LSTM) and deep belief network (DBN) models is utilized. Finally, the BOA is used to optimally select the hyperparameter values of the LSTM and DBN models. To elicit the enhanced performances of the BOA-EVCHAR technique, a series of experimentation analyses were performed. The extensive results of the BOA-EVCHAR technique show a superior value of 99.31% on the HAR process.

Defect Identification Method of Cable Termination based on Improved Gramian Angular Field and ResNet

Background: This paper proposes a defect identification method for vehicle-mounted cable terminals in electric multiple units (EMUs) based on the improved Graham angle field and residual network to address the issue of low recognition accuracy caused by the lack of partial discharge (PD) and identification data for Ethylene Propylene Rubber (EPR) cable terminal defects. Methods: The improved Gramian angular field (IGAF) characteristic transformation method was used to transform the PD one-dimensional time-series signal into a two-dimensional one after cable terminals with four common insulation defects were constructed, and a PD detection platform was built. Finally, an anti-aliasing downsampling module and attention mechanism were added to the residual network ResNet101 model. The Center loss and Softmax loss functions were integrated to increase accuracy for training and recognition classification. Topological feature images improved the distinguishability of defect categories. Results: The test results showed that the diagnostic method has an accuracy rate of 97.3% for identifying PD at the cable terminal. Conclusion: The proposed diagnosis model has higher recognition accuracy and better balance than other conventional fault diagnosis methods, making it suitable for diagnosing high-voltage cable faults in EMU trains.

Feature Extraction With Stacked Autoencoders for EEG Channel Reduction in Emotion Recognition.

Emotion recognition by electroencephalogram (EEG) signals is one of the complex methods because the extraction and recognition of the features hidden in the signal are sophisticated and require a significant number of EEG channels. Presenting a method for feature analysis and an algorithm for reducing the number of EEG channels fulfills the need for research in this field. Accordingly, this study investigates the possibility of utilizing deep learning to reduce the number of channels while maintaining the quality of the EEG signal. A stacked autoencoder network extracts optimal features for emotion classification in valence and arousal dimensions. Autoencoder networks can extract complex features to provide linear and non- linear features which are a good representative of the signal. The accuracy of a conventional emotion recognition classifier (support vector machine) using features extracted from SAEs was obtained at 75.7% for valence and 74.4% for arousal dimensions, respectively. Further analysis also illustrates that valence dimension detection with reduced EEG channels has a different composition of EEG channels compared to the arousal dimension. In addition, the number of channels is reduced from 32 to 12, which is an excellent development for designing a small-size EEG device by applying these optimal features.

Headland Identification and Ranging Method for Autonomous Agricultural Machines

Headland boundary identification and ranging are the key supporting technologies for the automatic driving of intelligent agricultural machinery, and they are also the basis for controlling operational behaviors such as autonomous turning and machine lifting. The complex, unstructured environments of farmland headlands render traditional image feature extraction methods less accurate and adaptable. This study utilizes deep learning and binocular vision technologies to develop a headland boundary identification and ranging system built upon the existing automatic guided tractor test platform. A headland image annotation dataset was constructed, and the MobileNetV3 network, notable for its compact model structure, was employed to achieve binary classification recognition of farmland and headland images. An improved MV3-DeeplabV3+ image segmentation network model, leveraging an attention mechanism, was constructed, achieving a high mean intersection over union (MIoU) value of 92.08% and enabling fast and accurate detection of headland boundaries. Following the detection of headland boundaries, binocular stereo vision technology was employed to measure the boundary distances. Field experiment results indicate that the system’s average relative errors of distance in ranging at distances of 25 m, 20 m, and 15 m are 6.72%, 4.80%, and 4.35%, respectively. This system is capable of meeting the real-time detection requirements for headland boundaries.

Exploration of machine learning models for surgical incision healing assessment based on thermal imaging: A feasibility study

AbstractIn this study, we explored the use of thermal imaging technology combined with computer vision techniques for assessing surgical incision healing. We processed 1189 thermal images, annotated by experts to define incision boundaries and healing statuses. Using these images, we developed a machine learning model based on YOLOV8, which automates the recognition of incision areas, lesion segmentation and healing classification. The dataset was divided into training, testing and validation sets in a 7:2:1 ratio. Our results show high accuracy rates in incision location recognition, lesion segmentation and healing classification, indicating the model's effectiveness as a precise and automated diagnostic tool for surgical incision healing assessment. Conclusively, our thermal image‐based machine learning model demonstrates excellent performance in wound assessment, paving the way for its clinical application in intelligent and standardized wound management.

Application of near-infrared spectral imaging and artificial intelligence classification in basketball motion image recognition

Application of near-infrared spectral imaging and artificial intelligence classification in basketball motion image recognition

A Novel Single-Word Speech Recognition on Embedded Systems Using a Convolution Neuron Network with Improved Out-of-Distribution Detection

Advancements in AI have elevated speech recognition, with convolutional neural networks (CNNs) proving effective in processing spectrogram-transformed speech signals. CNNs, with lower parameters and higher accuracy compared to traditional models, are particularly efficient for deployment on storage-limited embedded devices. Artificial neural networks excel in predicting inputs within their expected output range but struggle with anomalies. This is usually harmful to a speech recognition system. In this paper, the neural network classifier for speech recognition is trained with a “negative branch” method, incorporating directional regularization with out-of-distribution training data, allowing it to maintain a high confidence score to the input within distribution while expressing a low confidence score to the anomaly input. It can enhance the performance of anomaly detection of the classifier, addressing issues like misclassifying the speech command that is out of the distribution. The result of the experiment suggests that the accuracy of the CNN model will not be affected by the regularization of the “negative branch”, and the performance of abnormal detection will be improved as the number of kernels of the convolutional layer increases.

Refined composite multiscale dynamic causality diagram: a novel time-series analysis approach and its application in fault diagnosis of rotating machinery

Entropy-based feature extraction methods have been widely used in the fault diagnosis of rotating machinery, but the entropy-based methods still have the defects of poor noise robustness, weak feature extraction, and low computational efficiency. To solve this problem, this article proposes a fault diagnosis method based on refined composite multiscale dynamic causal diagram (RCMSDCD) and local receptive field extreme learning machine (LRFELM). First, a novel feature extraction method, named dynamic causal diagram (DCD), is proposed to comprehensively quantify static and dynamic complexity. DCD is obtained by combining generalized inverse fractional order entropy with complexity–entropy causal plane. Then, combined with the coarse-graining process, DCD is extended to a multiscale analysis called RCMSDCD to complement the feature description at cross-time scales. Third, RCMSDCD features are input into LRFELM classifier for fault recognition of rotating machinery. The effectiveness of the proposed RCMSDCD-LRFELM method is verified by the Paderborn University bearing test and real wind turbine gear signals. The results show that this method has the highest classification accuracy of 100% with high computational efficiency, good stability, and strong generalization ability.

Reconfigurable multifunctional neuromorphic memristor fabricated from two-dimensional ReSe2 ferroelectric nanosheet films

Memristor-based neuromorphic computing is beneficial for artificial intelligence to process external information autonomously with high speed and high efficiency. Two-dimensional (2D) layered van der Waals rhenium selenide (ReSe2) has optoelectronic and semiconductor properties, but its ferroelectricity has not been confirmed fully experimentally and the application exploration is currently limited. Here, we experimentally confirmed the room-temperature ferroelectricity of 2D ReSe2 and proposed a reconfigurable ReSe2 memristor that can realize multiple functions. The device can realize the conductance bidirectional regulation, and under the action of electrical signals, it exhibits the high 0.99 and 0.98 linearity and accurate bidirectional update of weights. Under the complementary effect of ReSe2 ferroelectric polarization flipping and interface defects, the device exhibits the memcapacitor and memristor reconfigurable behavior and multiple functions such as visible light perception, logical “OR” calculation, and long/short-term synaptic plasticity. In addition, the six-layer convolutional neural network built based on ReSe2 memristors can perform feature extraction and classification recognition of handwritten digital pictures, and its recognition accuracy can reach 97.04%. In addition to obtain substantial experimental evidence for the ferroelectricity of 2D ReSe2, this work also provides a new avenue for the implementation of ReSe2 ferroelectric memristors in the neuromorphic computing system with the front-end sensing and back-end processing.

Zero-exemplar deep continual learning for crop disease recognition: a study of total variation attention regularization in vision transformers.

With the increasing integration of AI technology in the food industry, deep learning has demonstrated its immense potential in the domain of plant disease image recognition. However, there remains a gap in research between models capable of continual learning of new diseases and addressing the inherent catastrophic forgetting issue in neural networks. This study aims to comprehensively evaluate various learning strategies based on advanced computer vision models for multi-disease continual learning tasks in food disease recognition. To cater to the benchmark dataset requirements, we collected the PlantDiseaseCL dataset, sourced from the internet, encompassing diverse crop diseases from apples, corn, and more. Utilizing the Vision Transformer (ViT) model, we established a plant disease image recognition classifier, which, in joint learning, outperformed several comparative CNN architectures in accuracy (0.9538), precision (0.9532), recall (0.9528), and F1 score (0.9560). To further harness the potential of ViT in food disease defect recognition, we introduced a mathematical paradigm for crop disease recognition continual learning. For the first time, we proposed a novel ViT-TV architecture in the multi-disease image recognition scenario, incorporating a Total Variation (TV) distance-based loss (TV-Loss) to quantify the disparity between current and previous attention distributions, fostering attention consistency and mitigating the catastrophic forgetting inherent in ViT without prior task samples. In the incremental learning of the PlantDiseaseCL dataset across 3-Steps and 5-Steps, our strategy achieved average accuracies of 0.7077 and 0.5661, respectively, surpassing all compared Zero-Exemplar Approaches like LUCIR, SI, MAS, and even outperforming exemplar-based strategies like EEIL and ICaRL. In conclusion, the ViT-TV approach offers robust support for the long-term intelligent development of the agricultural and food industry, especially showcasing significant applicability in continual learning for crop disease image recognition.

Recognition and (semi-)quantitative analysis of REE-bearing minerals in coal using automated scanning electron microscopy

The recognition and quantification of rare earth and yttrium (REY) minerals in coal is extremely important for understanding their geological origin and developing high-efficiency extraction techniques. This study developed an automated image analysis by scanning electron microscopy (AIA-SEM) method for characterizing REY minerals in coal. Carbon and copper were used as grey-scale calibration standard to assure the consistency per experiment and differentiate organic matter and minerals with different brightness. Two critical issues in determining REY including rapid recognition of fine REY-bearing particles and accurate classification and quantification of REY minerals in coal are resolved by (1) establishing grey threshold standard to locate REY-bearing grains and a two-step image scan; (2) developing classification criteria supplemented with spot- and EDS-mapping analysis for questionable particles. By using the established method, a total of 166 REY-bearing mineral grains were identified from 12,930 particles in a medium-volatile bituminous coal (REY concentration: 149.21 ppm; ash yield: 21.25%). It is found that Ca-REE flurocarbonates, monazite, and (Sc, Y)-bearing zircon are the dominant REY species in the studied coal sample. REE-aluminophosphates are also important hosts for REE in coal, with total amounts being >10%. REY minerals are fine in particle size. Over 50% of monazite, xenotime, Ca-REE flurocarbonate, and REE-aluminophosphate have particle size below 5 μm and monazite occurs mostly as submicron particle. Most REY minerals (> 60%) are embedded into coal macerals as included minerals. About 80% of REY-bearing phosphates and zircon are partly or completely encapsulated in clay minerals, whereas more than half of Ca-REY-fluorocarbonates mineral grains is free of other minerals.