Improve Recognition Performance Research Articles

A New Approach to Recognize Faces Amidst Challenges: Fusion Between the Opposite Frequencies of the Multi-Resolution Features

This paper proposes a new approach to pixel-level fusion using the opposite frequency from the discrete wavelet transform with Gaussian or Difference of Gaussian. The low-frequency from discrete wavelet transform sub-band was fused with the Difference of Gaussian, while the high-frequency sub-bands were fused with Gaussian. The final fusion was reconstructed using an inverse discrete wavelet transform into one enhanced reconstructed image. These enhanced images were utilized to improve recognition performance in the face recognition system. The proposed method was tested against benchmark face datasets such as The Database of Faces (AT&T), the Extended Yale B Face Dataset, the BeautyREC Face Dataset, and the FEI Face Dataset. The results showed that our proposed method was robust and accurate against challenges such as lighting conditions, facial expressions, head pose, 180-degree rotation of the face profile, dark images, acquisition with time gap, and conditions where the person uses attributes such as glasses. The proposed method is comparable to state-of-the-art methods and generates high recognition performance (more than 99% accuracy).

Building pattern recognition by using an edge-attention multi-head graph convolutional network

Effective building pattern recognition, a complex task that requires the simultaneous consideration of individual building features and spatial relations, is essential for successfully generalizing maps. However, existing deep learning approaches must still be adequately comprehensive in jointly quantifying the individual features and spatial relationships of buildings, suggesting further improvement in the quantitative representation of building spaces. This study presents a novel edge-attention multi-head graph convolutional network (GCN) that concurrently considers the quantitative modeling and representation of individual features and spatial relations, enhancing building pattern recognition. The proposed method captures individual building features and spatial relations, including proximity and arrangement similarity, by using spatial relationship descriptors and attention mechanisms to generate spatial relevance coefficients. These coefficients are then integrated into a weighted multi-head GCN to participate in the quantitative expression of individual features, facilitating the quantitative analysis and modeling of building features, and thus, improving recognition performance. Our experimental analysis confirms the method’s superior capability in recognizing complex spatial features. The method also demonstrates strong generalization across different scales and areas, underscoring its efficacy and potential for enhancing geospatial analyses.

Effects of a social network enhancement intervention for older adults: a feasibility study

BackgroundSocial networks play a critical role in the mental health of older adults. This pilot study investigates the feasibility of a newly developed intervention to enhance older adults’ social networks. This intervention was designed on the Theory of Mind’s foundation and aimed to enhance older adults’ social interaction motivation through theoretical explanations. Furthermore, the courses fostered more social opportunities for the participants through group-based sessions.MethodsThe feasibility of this intervention was tested using a double-blind, two-arm, non-randomized grouping approach. Older individuals residing in two separate residential buildings (n = 31, mean age = 66.81, 48% women) were divided into an intervention group (n = 15) and a control group (n = 16). They attended daily group sessions at a designated location and completed homework assignments. The primary outcomes of this pilot study were the feasibility of the intervention, and secondary outcomes included Theory of Mind levels and social network indicators. Additional outcomes encompassed levels of global mental health and depression.ResultsAll participants completed the pilot intervention and completed assessments. The primary outcomes indicated that the intervention had excellent feasibility, including compliance (attendance and homework completion rates met the standards) and satisfaction (average ratings for all items ranged from 4.47 to 5.00 on a 5-point scale). Interview results revealed that participants in the intervention group found the intervention beneficial for their daily lives and expressed a desire to participate in a formal intervention. Regarding secondary and additional outcomes, compared to the control group, the intervention group exhibited a significant improvement in emotional recognition performance of Theory of Mind. There was a significant increase in the whole network density in the intervention group. There were no significant differences in other social network indicators, global mental health, and depression levels in the intervention group compared to the control group.ConclusionsThe social network enhancement intervention for older adults is feasible. This pilot study has identified several improvements in the courses and tests. It is necessary to carry out a formal course to examine the effectiveness of the intervention on social networks in older adults.Trial registrationRegistration number: ChiCTR2100053779; Reg Date: 29/11/2021.

Optimization Study of Coal Gangue Detection in Intelligent Coal Selection Systems Based on the Improved Yolov8n Model

To address the low recognition accuracy of models for coal gangue images in intelligent coal preparation systems—especially in identifying small target coal gangue due to factors such as camera angle changes, low illumination, and motion blur—we propose an improved coal gangue separation model, Yolov8n-improvedGD(GD—Gangue Detection), based on Yolov8n. The optimization strategy includes integrating the GCBlock(Global Context Block) from GCNet(Global Context Network) into the backbone network to enhance the model’s ability to capture long-range dependencies in images and improve recognition performance. The CGFPN (Contextual Guidance Feature Pyramid Network) module is designed to optimize the feature fusion strategy and enhance the model’s feature expression capabilities. The GSConv-SlimNeck architecture is employed to optimize computational efficiency and enhance feature map fusion capabilities, thereby improving the model’s robustness. A 160 × 160 scale detection head is incorporated to enhance the sensitivity and accuracy of small coal and gangue detection, mitigate the effects of low-quality data, and improve target localization accuracy.

New Extracted Features to Recognize Faces Effected by Occlusions and Common Variations

Face recognition from non-identical face photos is a prominent area of research in pattern recognition and computer vision. Existing face recognition systems struggle with diverse changes like lighting conditions, expressions, and facial occlusions. This paper proposes a new Face Recognition (FR) approach that combines the Elastic Bunch Graph Matching (EBGM) approach with the greedy algorithm to automatically identify face landmarks. The proposed approach independently selects each optimal landmark of face image from different corresponding face images where the corresponding landmark of corresponding face image which achieves the best similarity is used rather than using one or at most two corresponding face images and computing the average between both. The locations of corresponding landmarks can be displaced to achieve maximum similarity with optimal landmarks. This proposed approach demonstrates improved recognition performance compared to contemporary face recognition methods. It effectively handles changing ratios of face parts and can recognize faces even with increasing occlusion sizes.

Intelligent Handwritten Identification Using Novel Hybrid Convolutional Neural Networks – Long-short-term Memory Architecture

Handwritten character identification finds broad applications in document analysis, digital forensics, and human-computer interaction. Conventional methods encounter challenges in accurately deciphering a range of handwriting styles and variations. Consequently, investigating intelligent system for handwriting identification becomes crucial to enhance accuracy and efficiency. This paper introduces an innovative hybrid deep learning architecture, seamlessly integrating the strengths of convolutional neural networks (CNN) and long-short term memory (LSTM) within a one single framework. The combination of these structures enables the model to effectively capture both spatial features and temporal dependencies inherent in handwritten strokes, resulting in improved recognition performance. The proposed 2DCNN-LSTM algorithm has been tested on MNIST dataset. The proposed hybrid CNN-LSTM structure has been compared to conventional intelligent machine learning methods, and the results demonstrate the superior performance of the hybrid CNN-LSTM, showcasing heightened accuracy, sensitivity, specificity, and other evaluation metrics.

Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks.

Understanding cognitive processes in the brain demands sophisticated models capable of replicating neural dynamics at large scales. We present a physiologically inspired speech recognition architecture, compatible and scalable with deep learning frameworks, and demonstrate that end-to-end gradient descent training leads to the emergence of neural oscillations in the central spiking neural network. Significant cross-frequency couplings, indicative of these oscillations, are measured within and across network layers during speech processing, whereas no such interactions are observed when handling background noise inputs. Furthermore, our findings highlight the crucial inhibitory role of feedback mechanisms, such as spike frequency adaptation and recurrent connections, in regulating and synchronizing neural activity to improve recognition performance. Overall, on top of developing our understanding of synchronization phenomena notably observed in the human auditory pathway, our architecture exhibits dynamic and efficient information processing, with relevance to neuromorphic technology.

Recognition of car horns based on principal component analysis of MEL frequency Cepstral coefficients and support vector machine

With the rapid increase in the number of automobiles, the noise pollution caused by car horns is attracting more attention. In order to accurately identify car horn sounds in traffic noise and strengthen traffic law enforcement, research on car horn sound recognition is becoming increasingly important. In this paper, using Principal Component Analysis (PCA) to optimize and reduce the dimensionality of extracted Mel Frequency Cepstral Coefficients (MFCC), feature vectors are input into four models: Support Vector Machine (SVM), Back Propagation (BP) Neural Network, Extreme Learning Machine (ELM), and Random Forest (RF). These models are utilized to classify and identify car horn sounds from traffic noise. The impact of feature dimensions and Signal-to-noise ratio(SNR) on recognition performance is analyzed. The results show that under this method, the recognition rates of the four models can be increased by increasing the number of Mel filters and feature dimensions in MFCC, which can improve recognition performance, especially when the SNR is lower. The SVM model exhibits the most significant improvement in recognition rate. They achieve the best recognition rates for measured car horn sounds of 98.66 %, 93.98 %, 89.44 %, and 95.12 %, respectively. Experimental results demonstrate that the proposed method has good recognition performance and can be effectively applied to the supervision of car horn usage on actual roads.

FlexibleCP: A data augmentation strategy for traffic sign detection

AbstractIn the field of traffic sign detection, effective data augmentation can improve the model's detection capacity, enabling the model to distinguish and locate traffic signs more precisely and enhancing driving safety. However, due to the small size and low representation of traffic signs in the dataset, standard common data augmentation techniques are not suitable for traffic sign detection. To address this issue, a novel data augmentation strategy called flexible cut and paste (FlexibleCP) is proposed. The overall enhancement approach is shifted from multi‐image fusion to target cropping and pasting. By introducing parameters to control the target pasting ratio and scaling ratio, the diversity of small target data and their size variations are enriched. Additionally, target size and type filters are added to enable targeted enhancement for different sizes and types of targets. This study, evaluates the proposed strategy using two representative traffic sign detection datasets, namely CTSD and GTSDB. The experimental results demonstrate a significant improvement in both detection and recognition performance of the model: on the CTSD dataset, the models trained with FlexibleCP data enhancement achieve 88.9% and 64.5% mAP0.5 and mAP0.5:0.95, respectively, which are 3.5% and 2.5% better than those trained with mosaic data enhancement; on the GTSDB dataset mAP0.5 and mAP0.5:0.95 reached 89.2% and 56.0%, respectively, an improvement of 4.0% and 3.9% over mosaic.

Advanced Multi-Label Fire Scene Image Classification via BiFormer, Domain-Adversarial Network and GCN

Detecting wildfires presents significant challenges due to the presence of various potential targets in fire imagery, such as smoke, vehicles, and people. To address these challenges, we propose a novel multi-label classification model based on BiFormer’s feature extraction method, which constructs sparse region-indexing relations and performs feature extraction only in key regions, thereby facilitating more effective capture of flame characteristics. Additionally, we introduce a feature screening method based on a domain-adversarial neural network (DANN) to minimize misclassification by accurately determining feature domains. Furthermore, a feature discrimination method utilizing a Graph Convolutional Network (GCN) is proposed, enabling the model to capture label correlations more effectively and improve performance by constructing a label correlation matrix. This model enhances cross-domain generalization capability and improves recognition performance in fire scenarios. In the experimental phase, we developed a comprehensive dataset by integrating multiple fire-related public datasets, and conducted detailed comparison and ablation experiments. Results from the tenfold cross-validation demonstrate that the proposed model significantly improves recognition of multi-labeled images in fire scenarios. Compared with the baseline model, the mAP increased by 4.426%, CP by 4.14% and CF1 by 7.04%.

Integration of multiscale fusion of residual neural network with 2-D gramian angular fields for lower limb movement recognition based on multi-channel sEMG signals

The human lower limb movements recognition (LLMR) plays a pivotal role in active lower limb exoskeleton robots. Employing surface electromyography (sEMG) signals for LLMR allows for the convenient, rapid and stable capture of signal variations, facilitating efficient identification of lower limb motion patterns. However, current sEMG-based LLMR methods face challenges such as incomplete feature extraction, limited contextual information and restricted feature extraction scales during feature extraction. This paper proposed a LLMR method based on Gramian Angular Fields (GAF) and multiscale fusion of Residual Neural Network (MS-ResNet). The denoised sEMG time series was transformed into Gramian Angular Difference Field (GADF) matrix based on GAF. The MS-ResNet model, incorporating ResNet and multiscale feature fusion concepts, was proposed to comprehensively capture global and local information through different-scale feature extraction and fusion, so as to improve recognition performance. sEMG signals from 11 muscles of the preferred leg of 15 healthy subjects were recorded during six common lower limb movements. Experimental analysis investigated the impact of the convolutional kernel size (k × k) in Stream 2 of MS-ResNet and the number of muscles involved on recognition performance. The study revealed that selecting k as 13, coupled with 11 muscles, yielded optimal model performance with the average cross-individual recognition accuracy reaching 97.62 %, demonstrating the model’s efficiency in LLMR. This method could provide a viable solution for developing more efficient and reliable LLMR systems, applicable to lower limb exoskeleton robots and intelligent prosthetics.

A non-local based microcrack segmentation model optimized for effective high resolution and low-power devices

Concrete is notable for its strength, durability, and versatility, making it a fundamental material in construction. Nevertheless, microcracks occur in concrete structures due to various factors, and the size of these cracks gradually widens over time, which can lead to spalling. Furthermore, the advent of climate change accelerates concrete degradation, posing significant challenges to structural maintenance. Recently, deep learning-based computer vision techniques, especially convolutional neural networks, have shown promising results in detecting and segmenting cracks in concrete walls. In this study, we propose a two-step training strategy for a microcrack segmentation model optimized for high resolution and low-power devices. The first step is to develop a segmentation model that can detect microcracks through high-resolution images. This strategy includes generating a high-quality CrackHQ dataset using super-resolution and a Transformer-based model training process named CegFormer. The second step involves converting the trained model to a lightweight version, increasing inference speed and making it operable on low-power devices. While performing lightweight, the model performance inevitably is degraded. We restored the degraded crack recognition performance by applying knowledge distillation. The experimental results of comparing baseline and CegFormer showed a 12.82% improvement in microcrack recognition performance. The detector optimized for use in actual fields improved recognition performance by up to 5.16% and reduced inference speed by 51.52%. The utilization of the two-step training strategy enabled the optical sensors to accurately detect microcracks even from far fields. Future work will focus on developing crack detectors that take into account both indoor and outdoor environmental conditions, as well as different texture of surface materials, for application in building inspection.

Biomemristor Reservoir Computing With Multi-Value Mask for Improving Recognition Performance

Biomemristor Reservoir Computing With Multi-Value Mask for Improving Recognition Performance

Synthetic lidar point cloud generation using deep generative models for improved driving scene object recognition

The imbalanced distribution of different object categories poses a challenge for training accurate object recognition models in driving scenes. Supervised machine learning models trained on imbalanced data are biased and easily overfit the majority classes, such as vehicles and pedestrians, which appear more frequently in driving scenes. We propose a novel data augmentation approach for object recognition in lidar point cloud of driving scenes, which leverages probabilistic generative models to produce synthetic point clouds for the minority classes and complement the original imbalanced dataset. We evaluate five generative models based on different statistical principles, including Gaussian mixture model, variational autoencoder, generative adversarial network, adversarial autoencoder and the diffusion model. Experiments with a real-world autonomous driving dataset show that the synthetic point clouds generated for the minority classes by the Latent Generative Adversarial Network result in significant improvement of object recognition performance for both minority and majority classes. The codes are available at https://github.com/AAAALEX-XIANG/Synthetic-Lidar-Generation.

SFT-Net: A Network for Detecting Fatigue From EEG Signals by Combining 4D Feature Flow and Attention Mechanism.

Fatigued driving is a leading cause of traffic accidents, and accurately predicting driver fatigue can significantly reduce their occurrence. However, modern fatigue detection models based on neural networks often face challenges such as poor interpretability and insufficient input feature dimensions. This article proposes a novel Spatial-Frequency-Temporal Network (SFT-Net) method for detecting driver fatigue using electroencephalogram (EEG) data. Our approach integrates EEG signals' spatial, frequency, and temporal information to improve recognition performance. We transform the differential entropy of five frequency bands of EEG signals into a 4D feature tensor to preserve these three types of information. An attention module is then used to recalibrate the spatial and frequency information of each input 4D feature tensor time slice. The output of this module is fed into a depthwise separable convolution (DSC) module, which extracts spatial and frequency features after attention fusion. Finally, long short-term memory (LSTM) is used to extract the temporal dependence of the sequence, and the final features are output through a linear layer. We validate the effectiveness of our model on the SEED-VIG dataset, and experimental results demonstrate that SFT-Net outperforms other popular models for EEG fatigue detection. Interpretability analysis supports the claim that our model has a certain level of interpretability. Our work addresses the challenge of detecting driver fatigue from EEG data and highlights the importance of integrating spatial, frequency, and temporal information.

Recognition of aggressive driving behavior under abnormal weather based on Convolutional Neural Network and transfer learning

Objectives Aggressive driving behavior can lead to potential traffic collision risks, and abnormal weather conditions can exacerbate this behavior. This study aims to develop recognition models for aggressive driving under various climate conditions, addressing the challenge of collecting sufficient data in abnormal weather. Methods Driving data was collected in a virtual environment using a driving simulator under both normal and abnormal weather conditions. A model was trained on data from normal weather (source domain) and then transferred to foggy and rainy weather conditions (target domains) for retraining and fine-tuning. The K-means algorithm clustered driving behavior instances into three styles: aggressive, normal, and cautious. These clusters were used as labels for each instance in training a CNN model. The pre-trained CNN model was then transferred and fine-tuned for abnormal weather conditions. Results The transferred models showed improved recognition performance, achieving an accuracy score of 0.81 in both foggy and rainy weather conditions. This surpassed the non-transferred models’ accuracy scores of 0.72 and 0.69, respectively. Conclusions The study demonstrates the significant application value of transfer learning in recognizing aggressive driving behaviors with limited data. It also highlights the feasibility of using this approach to address the challenges of driving behavior recognition under abnormal weather conditions.

Deep migration learning-based recognition of diseases and insect pests in Yunnan tea under complex environments

BackgroundThe occurrence, development, and outbreak of tea diseases and pests pose a significant challenge to the quality and yield of tea, necessitating prompt identification and control measures. Given the vast array of tea diseases and pests, coupled with the intricacies of the tea planting environment, accurate and rapid diagnosis remains elusive. In addressing this issue, the present study investigates the utilization of transfer learning convolution neural networks for the identification of tea diseases and pests. Our objective is to facilitate the accurate and expeditious detection of diseases and pests affecting the Yunnan Big leaf kind of tea within its complex ecological niche.ResultsInitially, we gathered 1878 image data encompassing 10 prevalent types of tea diseases and pests from complex environments within tea plantations, compiling a comprehensive dataset. Additionally, we employed data augmentation techniques to enrich the sample diversity. Leveraging the ImageNet pre-trained model, we conducted a comprehensive evaluation and identified the Xception architecture as the most effective model. Notably, the integration of an attention mechanism within the Xeption model did not yield improvements in recognition performance. Subsequently, through transfer learning and the freezing core strategy, we achieved a test accuracy rate of 98.58% and a verification accuracy rate of 98.2310%.ConclusionsThese outcomes signify a significant stride towards accurate and timely detection, holding promise for enhancing the sustainability and productivity of Yunnan tea. Our findings provide a theoretical foundation and technical guidance for the development of online detection technologies for tea diseases and pests in Yunnan.

SAM-CFFNet: SAM-Based Cross-Feature Fusion Network for Intelligent Identification of Landslides

Landslides are common hazardous geological events, and accurate and efficient landslide identification methods are important for hazard assessment and post-disaster response to geological disasters. Deep learning (DL) methods based on remote sensing data are currently widely used in landslide identification tasks. The recently proposed segment anything model (SAM) has shown strong generalization capabilities in zero-shot semantic segmentation. Nevertheless, SAM heavily relies on user-provided prompts, and performs poorly in identifying landslides on remote sensing images. In this study, we propose a SAM-based cross-feature fusion network (SAM-CFFNet) for the landslide identification task. The model utilizes SAM’s image encoder to extract multi-level features and our proposed cross-feature fusion decoder (CFFD) to generate high-precision segmentation results. The CFFD enhances landslide information through fine-tuning and cross-fusing multi-level features while leveraging a shallow feature extractor (SFE) to supplement texture details and improve recognition performance. SAM-CFFNet achieves high-precision landslide identification without the need for prompts while retaining SAM’s robust feature extraction capabilities. Experimental results on three open-source landslide datasets show that SAM-CFFNet outperformed other comparative models in terms of landslide identification accuracy and achieved an intersection over union (IoU) of 77.13%, 55.26%, and 73.87% on the three datasets, respectively. Our ablation studies confirm the effectiveness of each module designed in our model. Moreover, we validated the justification for our CFFD design through comparative analysis with diverse decoders. SAM-CFFNet achieves precise landslide identification using remote sensing images, demonstrating the potential application of the SAM-based model in geohazard analysis.

An Enhanced Human Gait Recognition System using UALNN Classifier

Human gait recognition is a biometric identification method that uses a person’s gait patterns to identify them. Gait Recognition (GR) is the steps to follow in recognizing walking style of an individual. It identifies a person as of a longer distance with higher precision owing to its unique characteristics; thus, it is mostly utilized for tracking purposes. In computer vision software, specifically, to track the same individual across multiple scenarios, the individual recognition across numerous cameras is a significant task. However, owing to the position of caught image or video, human identification by means of GR could be limited. Therefore, for person re-identification, a novel GR methodology has been proposed here. Five phases were undergone by the proposed methodology. Initially, from an openly accessible dataset, the input video is collected, which is then transmuted into a number of frames. To eliminate the noise from the frames, the Mean Modified Filter (MMF) is utilized after frame conversion. The Gaussian Taxicab Mixture Model (GTMM) is employed to get rid of the background from the frames. After that, Similarity Histogram Silhouette Segmentation (SHSS) model is utilized to segment the humans’ silhouette as of the background removed frames; thus, generating the energy image. The features are extracted from this energy image. The Uniform Alex Neural Network classifier (UALNN) uses these features to identify the human gait. Improved recognition performance is attained from the experiential outcomes. The overall recognition outcomes obtained concluded that the proposed methodology is highly promising.

Multimodal fine-grained grocery product recognition using image and OCR text

Automatic recognition of grocery products can be used to improve customer flow at checkouts and reduce labor costs and store losses. Product recognition is, however, a challenging task for machine learning-based solutions due to the large number of products and their variations in appearance. In this work, we tackle the challenge of fine-grained product recognition by first extracting a large dataset from a grocery store containing products that are only differentiable by subtle details. Then, we propose a multimodal product recognition approach that uses product images with extracted OCR text from packages to improve fine-grained recognition of grocery products. We evaluate several image and text models separately and then combine them using different multimodal models of varying complexities. The results show that image and textual information complement each other in multimodal models and enable a classifier with greater recognition performance than unimodal models, especially when the number of training samples is limited. Therefore, this approach is suitable for many different scenarios in which product recognition is used to further improve recognition performance. The dataset can be found at https://github.com/Tubbias/finegrainocr.