Robust Recognition Research Articles

Intelligent road surface state recognition method based on multi-layer attention residual network

Abstract Data-driven road surface state recognition enhances the efficiency and accuracy of road management, contributing to increased safety and reliability in road traffic. However, traditional machine learning and deep learning-based road surface state recognition typically rely on extensive data for model training, making it challenging to adapt to complex tasks in diverse scenarios. Therefore, this paper proposes a Multi-layer Attention Residual Network (MARN)-based intelligent road surface state recognition method. First, a Residual Convolutional Neural Network (ResNet) is constructed as the backbone model of MARN to mitigate the gradient vanishing problem, allowing the network to extract deeper features. Subsequently, an adaptive multi-layer attention mechanism is introduced in each convolutional layer, enabling adaptive weighting of each feature channel in the dataset to enhance the model’s focus on different features for better feature extraction. Furthermore, a cosine annealing learning rate adjuster is designed to improve the accuracy, robustness, and convergence during the model training process. Finally, the proposed MARN is validated using an image dataset containing six different road surface states. Comparative studies are conducted on the recognition accuracy of the proposed MARN, original ResNet, Visual Geometry Group network (VGG16), and Convolutional Neural Network (CNN). The impact of different batch sizes on the convergence speed of road surface state recognition under MARN is also analyzed. Results demonstrate that MARN achieves a training set accuracy of over 95%, surpassing VGG16 and CNN with accuracies below 85%. Compared to ResNet, MARN exhibits a 1.3% higher training set accuracy and a 0.25 lower validation set loss, showcasing superior accuracy and robustness in road surface state recognition.

Design of automatic sorting system based on capsule neural network

Abstract. This paper proposes a vegetable disease classification system based on Capsule Neural Network (CapsNet), which uses the visual recognition mechanism of deep learning algorithms to improve the efficiency and accuracy of agricultural disease recognition. The CapsNet algorithm outperforms traditional Convolutional Neural Network (CNN) in few-shot learning and generalization due to its unique capsule structure and dynamic routing mechanism. Experimental results show that the proposed system performs well in the accuracy, robustness, and model interpretation of image recognition, and can accurately identify diseases even in the case of poor image quality or noise and occlusion. In addition, the embedded system of this experiment includes an image acquisition module, an algorithm core module, and a logic control module, which realizes the automation of the whole process from image acquisition to recognition result display. This study also discusses the scalability and adaptability of the system, as well as the future application prospects in agricultural intelligence and automation. It is expected that CapsNet will play a more critical role in artificial intelligence, especially in agrarian automation and smart agriculture.

Bioinspired Hybrid and Incomplete Data Clustering

Enhancing the results of data clustering for hybrid (numeric and categorical) and missing data is paramount for robust pattern recognition and decision-making in various fields. Traditional clustering algorithms often need help with heterogeneous data types and incomplete information, leading to suboptimal groupings and potentially biased insights. By addressing these challenges, advanced techniques, such as bioinspired algorithms, can provide more accurate and comprehensive clustering. We show the usefulness of using internal validity indices in obtaining high-quality clusters for hybrid and incomplete data. In addition, this work analyzes the influence of applying the recently proposed PAntSA on the performance of hybrid and incomplete clustering algorithms. To do this, the results of different algorithms are compared before and after applying PAntSA. The statistical analysis of the results provides experimental evidence that supports that the PAntSA algorithm improves the quality of the groups obtained by hybrid and incomplete data clustering methods.

Automated production of batched unclonable micro-patterns anti-counterfeiting labels with strong robustness and rapid recognition speed

Anti-counterfeiting technologies are indeed crucial for information security and protecting product authenticity. Traditional anti-counterfeiting methods have their limitations due to their clonable nature. Exploring new technologies, particularly those based on pixel-level textures, is a promising avenue to address the clonable issue due to high encoding capacity. However, research in this field is still in its early stages. This work introduces a new fluorescent anti-counterfeiting label technology with four key characteristics: efficient laser etching, high-throughput fabrication and segmentation, robustness aided by data augmentation, and an exceptionally high recognition speed. Specifically, the etching achieves a speed of 1,200 labels/3s, the high-throughput procedures yield a rate of 2,400 labels/4 mins, and the total count is about 5.2 × 104 labels. The number of labels is further augmented to about 5.2 × 106 by implementing arbitrary rotation and brightness variation to enhance the robustness in the recognition procedure. We divide these labels into 44 categories based on differences in patterns. Utilizing machine learning methods, we have achieved a total recognition (including extraction and search process) time per label averaging 421.96 ms without classification, and 40.13 ms with classification. Specifically, the search process with classification is nearly fiftieth times shorter than the non-classification method, reaching 8.52 ms in average. The overall recognition time is much faster than previous works, and achieves an accuracy of over 98.7%. This work significantly increases the practical significance of pixel-level anti-counterfeiting labels.

An Analysis of Critical Processes and Employee Challenges at Velmurugan Heavy Engineering Industries (VHEI) Private Limited, Thanjavur

This study investigates the critical processes and challenges faced by employees at Velmurugan Heavy Engineering Pvt. Ltd., located in Sengipatti, Thanjavur. The aim is to identify areas for improvement and provide recommendations to enhance employee performance and organizational effectiveness. Employing a mixed-method approach combining qualitative and quantitative methods, data were collected extensively through surveys, interviews, and observations from a representative sample across departments and hierarchical levels. The findings highlight critical areas such as project management, quality control, communication channels, and performance evaluation. While the organization demonstrates competence in these areas, gaps exist, particularly in streamlined communication and efficient project management. The study also uncovers challenges including unclear role expectations, limited professional development opportunities, inadequate work-life balance, and insufficient recognition and rewards. Recommendations are offered to strengthen communication channels, enhance project management practices, implement employee development programs, promote work-life balance initiatives, and establish a robust recognition and rewards system. Addressing these challenges and implementing recommended strategies has led to improved employee satisfaction, productivity, and overall organizational performance. This study provides valuable insights for Velmurugan Heavy Engineering Pvt. Ltd. to cultivate a positive work environment and foster a highly engaged workforce, contributing to sustainable growth and success

An estimation method for vision-based autonomous landing system for fixed wing aircraft

In this paper, we propose a vision-based estimation method for autonomous landing of fixed-wing aircraft for Advanced Air Mobility. The concept of autonomous flight with minimal human intervention has gained significant interest with a particular focus on autonomous landing. The goal is to overcome the limitations of existing instrument landing systems and reduce reliance on GNSS during aircraft landing. The proposed system leverages the following techniques for high-performance and real-time operations in various real-world environments and during all stages of landing. A deep learning segmentation model was directly learned, created, and used for robust runway recognition performance against environmental changes around the runway. Algorithms were designed to adapt to different flight stages considering the varying information offered by image data when the aircraft is in mid-air and on the ground. The relative lateral position from the aircraft to the runway was estimated using bird’s-eye-view conversion and inverse projection techniques instead of conventional perspective-n-point methods for reducing the recognition error occurring from the conversion between 2D image and 3D world. Finally, the mixed-precision technique was used for the segmentation model to improve inference speed and ensure real-time operation in real-world deployment. Estimation stability and reliability were improved by using a Kalman Filter to cope with sensor uncertainties caused by the real-world flight environment. Consequently, we show that the average error in the lateral position estimation by the proposed method is comparable to the accuracy of a single GNSS and demonstrate successful autonomous landing of our test aircraft with a set of flight tests.

Attendance Marking using Face Detection

The demand for automated attendance management systems has significantly increased in educational institutions and corporate environments where manual methods are prone to inefficiencies and errors. This paper presents a real-time, facial recognition-based attendance system developed using Flask, OpenCV, and machine learning techniques. The system employs OpenCV’s Haar Cascade classifier for face detection and the K-Nearest Neighbours (KNN) algorithm for face recognition. During user registration, the system captures 10 images per individual to ensure robust recognition across various facial expressions. The web interface, built using Flask, facilitates user interaction for managing attendance records stored in CSV format. The system achieves an accuracy of 92% under optimal lighting conditions, though performance decreases to 80% in low-light environments. Recognition is completed in approximately 2 seconds, making the system suitable for real-time applications in classrooms or offices. Despite handling real-time image streaming efficiently, challenges such as reduced accuracy when handling obstructions like masks or glasses remain. Future enhancements will explore the integration of deep learning models, such as Convolutional Neural Networks (CNNs), to improve robustness and scalability. This system demonstrates the potential of facial recognition technology for automating attendance tracking, with significant applications in both educational and corporate sectors

Digital signal detection and recognition in the communication field combining DAEN and CNN

In communication, the detection and recognition of digital signals have always faced problems such as low adaptability and high misclassification rate. To address these issues, this study innovatively fused the deep auto-encoder network with convolutional neural networks to construct a novel digital signal detection and recognition model. Firstly, this study utilized the powerful feature extraction capabilities of the deep auto-encoder network to extract key feature information from massive amounts of data. Then these features were combined with convolutional neural networks to construct a detection and recognition model. These results confirmed that the constructed digital signal detection and recognition model had values of 93.75% and 94.18% in data detection recognition rate and average classification accuracy, respectively. Meanwhile, this model also performed well in terms of data processing accuracy and recall. In the comparison of data processing, the accuracy and recall rates were 93.59% and 94.67%, respectively, and the performance of data detection and recognition was better than that of the comparison methods. This indicates that the constructed digital signal detection and recognition model can significantly improve the reliability and robustness of signal detection and recognition. This paper brings new breakthroughs to the development of digital signal detection and recognition technology in communication.

A Two-Stream Method for Human Action Recognition Using Facial Action Cues.

Human action recognition (HAR) is a critical area in computer vision with wide-ranging applications, including video surveillance, healthcare monitoring, and abnormal behavior detection. Current HAR methods predominantly rely on full-body data, which can limit their effectiveness in real-world scenarios where occlusion is common. In such situations, the face often remains visible, providing valuable cues for action recognition. This paper introduces Face in Action (FIA), a novel two-stream method that leverages facial action cues for robust action recognition under conditions of significant occlusion. FIA consists of an RGB stream and a landmark stream. The RGB stream processes facial image sequences using a fine-spatio-multitemporal (FSM) 3D convolution module, which employs smaller spatial receptive fields to capture detailed local facial movements and larger temporal receptive fields to model broader temporal dynamics. The landmark stream processes facial landmark sequences using a normalized temporal attention (NTA) module within an NTA-GCN block, enhancing the detection of key facial frames and improving overall recognition accuracy. We validate the effectiveness of FIA using the NTU RGB+D and NTU RGB+D 120 datasets, focusing on action categories related to medical conditions. Our experiments demonstrate that FIA significantly outperforms existing methods in scenarios with extensive occlusion, highlighting its potential for practical applications in surveillance and healthcare settings.

A Student Facial Expression Recognition Model Based on Multi-Scale and Deep Fine-Grained Feature Attention Enhancement.

In smart classroom environments, accurately recognizing students' facial expressions is crucial for teachers to efficiently assess students' learning states, timely adjust teaching strategies, and enhance teaching quality and effectiveness. In this paper, we propose a student facial expression recognition model based on multi-scale and deep fine-grained feature attention enhancement (SFER-MDFAE) to address the issues of inaccurate facial feature extraction and poor robustness of facial expression recognition in smart classroom scenarios. Firstly, we construct a novel multi-scale dual-pooling feature aggregation module to capture and fuse facial information at different scales, thereby obtaining a comprehensive representation of key facial features; secondly, we design a key region-oriented attention mechanism to focus more on the nuances of facial expressions, further enhancing the representation of multi-scale deep fine-grained feature; finally, the fusion of multi-scale and deep fine-grained attention-enhanced features is used to obtain richer and more accurate facial key information and realize accurate facial expression recognition. The experimental results demonstrate that the proposed SFER-MDFAE outperforms the existing state-of-the-art methods, achieving an accuracy of 76.18% on FER2013, 92.75% on FERPlus, 92.93% on RAF-DB, 67.86% on AffectNet, and 93.74% on the real smart classroom facial expression dataset (SCFED). These results validate the effectiveness of the proposed method.

Dual-Feature Fusion Learning: An Acoustic Signal Recognition Method for Marine Mammals

Marine mammal acoustic signal recognition is a key technology for species conservation and ecological environment monitoring. Aiming at the complex and changing marine environment, and because the traditional recognition method based on a single feature input has the problems of poor environmental adaptability and low recognition accuracy, this paper proposes a dual-feature fusion learning method. First, dual-domain feature extraction is performed on marine mammal acoustic signals to overcome the limitations of single feature input methods by interacting feature information between the time-frequency domain and the Delay-Doppler domain. Second, this paper constructs a dual-feature fusion learning target recognition model, which improves the generalization ability and robustness of mammal acoustic signal recognition in complex marine environments. Finally, the feasibility and effectiveness of the dual-feature fusion learning target recognition model are verified in this study by using the acoustic datasets of three marine mammals, namely, the Fraser’s Dolphin, the Spinner Dolphin, and the Long-Finned Pilot Whale. The dual-feature fusion learning target recognition model improved the accuracy of the training set by 3% to 6% and 20% to 23%, and the accuracy of the test set by 1% to 3% and 25% to 38%, respectively, compared to the model that used the time-frequency domain features and the Delay-Doppler domain features alone for recognition.

The Research of Intra-Pulse Modulated Signal Recognition of Radar Emitter under Few-Shot Learning Condition Based on Multimodal Fusion

Radar radiation source recognition is critical for the reliable operation of radar communication systems. However, in increasingly complex electromagnetic environments, traditional identification methods face significant limitations. These methods often struggle with high noise levels and diverse modulation types, making it difficult to maintain accuracy, especially when the Signal-to-Noise Ratio (SNR) is low or the available training data are limited. These difficulties are further intensified by the necessity to generalize in environments characterized by a substantial quantity of noisy, low-quality signal samples while being constrained by a limited number of desirable high-quality training samples. To more effectively address these issues, this paper proposes a novel approach utilizing Model-Agnostic Meta-Learning (MAML) to enhance model adaptability in few-shot learning scenarios, allowing the model to quickly learn with limited data and optimize parameters effectively. Furthermore, a multimodal fusion neural network, DCFANet, is designed, incorporating residual blocks, squeeze and excitation blocks, and a multi-scale CNN, to fuse I/Q waveform data and time–frequency image data for more comprehensive feature extraction. Our model enables more robust signal recognition, even when the signal quality is severely degraded by noise or when only a few examples of a signal type are available. Testing on 13 intra-pulse modulated signals in an Additive White Gaussian Noise (AWGN) environment across SNRs ranging from −20 to 10 dB demonstrated the approach’s effectiveness. Particularly, under a 5−way5−shot setting, the model achieves high classification accuracy even at −10dB SNR. Our research underscores the model’s ability to address the key challenges of radar emitter signal recognition in low-SNR and data-scarce conditions, demonstrating its strong adaptability and effectiveness in complex, real-world electromagnetic environments.

Multi-resolution visual Mamba with multi-directional selective mechanism for retinal disease detection.

Retinal diseases significantly impact patients' quality of life and increase social medical costs. Optical coherence tomography (OCT) offers high-resolution imaging for precise detection and monitoring of these conditions. While deep learning techniques have been employed to extract features from OCT images for classification, convolutional neural networks (CNNs) often fail to capture global context due to their focus on local receptive fields. Transformer-based methods, on the other hand, suffer from quadratic complexity when handling long-range dependencies. To overcome these limitations, we introduce the Multi-Resolution Visual Mamba (MRVM) model, which addresses long-range dependencies with linear computational complexity for OCT image classification. The MRVM model initially employs convolution to extract local features and subsequently utilizes the retinal Mamba to capture global dependencies. By integrating multi-scale global features, the MRVM enhances classification accuracy and overall performance. Additionally, the multi-directional selection mechanism (MSM) within the retinal Mamba improves feature extraction by concentrating on various directions, thereby better capturing complex, orientation-specific retinal patterns. Experimental results demonstrate that the MRVM model excels in differentiating retinal images with various lesions, achieving superior detection accuracy compared to traditional methods, with overall accuracies of 98.98\% and 96.21\% on two public datasets, respectively. This approach offers a novel perspective for accurately identifying retinal diseases and could contribute to the development of more robust artificial intelligence algorithms and recognition systems for medical image-assisted diagnosis.

Enhancing image recognition robustness in early weed detection through optimal training data curation

Abstract Using convolutional neural networks (CNNs) for image recognition is effective for early weed detection. However, the impact of training data curation, specifically concerning morphological changes during the early growth phases of weeds, on recognition robustness remains unclear. We focused on four weed species (giant ragweed [Ambrosia trifida L.], red morningglory [Ipomoea coccinea L.], pitted morningglory [Ipomoea lacunosa L.], and burcucumber [Sicyos angulatus L.]) with varying cotyledon and true leaf shapes. Creating 16 models in total, we employed four dataset patterns with different growth stage combinations, two image recognition algorithms (object detection: You Look Only Once [YOLO] v5 and image classification: Visual Geometry Group [VGG] 19), and two conditions regarding the number of species treated (four and two species). We evaluated the effects of growth stage training on weed recognition success using two datasets. One evaluation revealed superior results with a single class/species training dataset, achieving >90% average precision for detection and classification accuracy under most conditions. The other dataset revealed that merging different growth stages with different shapes as a class effectively prevented misrecognition among different species when using YOLOv5. Both results suggest that integrating different shapes in a plant species as a single class is effective for maintaining robust recognition success amid temporal morphological changes during the early growth stage. This finding not only enhances early detection of weed seedlings but also bolsters the robustness of general plant species identification.

Research on fiber OTDR signal recognition and classification based on the ViT model

In the field of event recognition for phase-sensitive optical time-domain reflectometry (-OTDR), convolutional neural networks (CNNs) have been the mainstream tool. However, Transformer models, with their self-attention mechanism, have provided a new perspective for image recognition tasks. This paper proposes a -OTDR event recognition method based on the Transformer model and experimentally demonstrates its significant advantages over CNNs. The method utilizes an attention-based feature extraction approach to recognize distributed fiber optic sensing signals in a phase-sensitive optical time-domain reflectometer (-OTDR). In perimeter defense applications, various interferences and noise severely affect the recognition of -OTDR sensing signals, leading to false alarms, making accurate signal recognition both challenging and necessary. Extracting signal features and using machine learning classification models to recognize signals has been a research hotspot in recent years. The effectiveness of feature extraction is crucial to classification performance. This project introduces a temporal-spatial dual attention mechanism into the feature extraction of fiber optic sensing signals to improve the accuracy and robustness of signal recognition.

Innovative strategies for intelligent services in smart libraries in the information age based on linear discriminant analysis

With the advent of the information age, to provide better services and ensure the security management of libraries, intelligent facial recognition technology has gradually become a hot research direction in library management. Meanwhile, to further improve the comprehensive performance of facial recognition, this study attempts to integrate principal component analysis and linear discriminant analysis on the basis of analyzing the framework of recognition technology. Afterwards, it introduced support vector machines for recognition and classification, and proposed a new recognition model. The experimental results show that the recognition accuracy of the proposed model is up to 97 % in the ORL dataset and 94 % in the Yale dataset. The recognition error rate is as low as 0.1 % when the number of training samples is 215 and the number of iterations is 200. The model has the best recognition performance when the image size is 25 × 25 mm and the number of noises is 10. In addition, the model is particularly effective in recognition on single person color or gray images, with the highest P-value of 98.7 %, the highest R-value of 98.8 %, and the highest F1-value of 97.5 %. These results show that the proposed model significantly improves the accuracy and robustness of face recognition, and provides strong technical support for intelligent service innovation in smart libraries.

Tetrahedral DNA-Based Ternary Recognition Ratiometric Fluorescent Probes for Real-Time In Situ Resolving Lysosome Subpopulations in Living Cells via Cl-, Ca2+, and pH.

Lysosomes are multifunctional organelles vital for cellular homeostasis with distinct subpopulations characterized by varying levels of Cl-, Ca2+, and H+. In situ visualization of these parameters is crucial for lysosomal research, yet developing probes that can simultaneously detect multiple ions remains challenging. Herein, we developed a lysosome-targeting ternary recognition ratiometric fluorescent probe based on tetrahedral DNA nanostructures (TDNs) to analyze lysosome subpopulations by Cl-, Ca2+, and pH. The TDN probe is assembled from four single-stranded DNAs, each end-modified with responsive fluorophores (Pr-Cl for Cl-, Pr-Ca for Ca2+, and Pr-pH for pH) or a reference fluorophore (Cy5). The fluorophores are integrated at the vertices of the rigid TDN to minimize mutual interference, and their fixed stoichiometry establishes a robust ternary recognition ratiometric fluorescence sensor for in situ resolution of lysosome subpopulations in living cells. Accordingly, a rise in lysosome subpopulations 2/6 characterized by low [Cl-], medium/high [Ca2+], and high pH was observed in the Niemann-Pick disease model cells but seldom observed in the control group. Conversely, there was a marked decline in the fraction of subpopulations 1/4/5 characterized by high [Cl-], medium to low [Ca2+], and pH. These changes were substantially reversed upon treatment. The probe holds great promise for studying lysosome subpopulations and the diagnosis and treatment of related diseases.

A Novel Non-Iterative Deep Convolutional Neural Network with Kernelized Classification for Robust Face Recognition

A Novel Non-Iterative Deep Convolutional Neural Network with Kernelized Classification for Robust Face Recognition

Analysing the robustness of finger vein recognition: cross-dataset reliability and vein utility

AbstractFinger vein recognition is an emerging biometric trait known for its privacy features. Despite the remarkable performance of deep learning methods like convolutional neural networks on challenging finger vein datasets, their reliability and robustness need further examination. This study evaluates the robustness of three recognition methods—the traditional Miura Method, a supervised convolutional neural network, and an unsupervised convolutional auto-encoder—through the challenging and more realistic scenario of cross-dataset comparisons. We also analyse the reliability of these methods in terms of sample quality. We introduce a novel vein quality metric to measure vein clarity and complexity and compare it against an existing image quality metric, natural image quality evaluator. Our findings reveal differences in how these recognition methods utilise finger vein images for comparisons, highlighting the need for robust recognition techniques in more realistic scenarios. In addition, our vein quality metric effectively detects defective images, reducing the zero false-match rate from 34.98% to 8.18% on the SDUMLA-HMT dataset. These results indicate the need for metrics more focussing on finger vein image characteristics for effective quality assessment for finger vein images.

Deep Learning for Robust Facial Expression Recognition: A Resilient Defense Against Adversarial Attacks

Adversarial attacks can be extremely dangerous, particularly in scenarios where the precision of facial expression identification is of utmost importance. Hiring adversarial training methods proves effective in mitigating these threats. Although effective, this technique requires large computing resources. This study aims to strengthen deep learning model resilience against adversarial attacks while optimizing performance and resource efficiency. Our proposed method uses adversarial training techniques to create adversarial examples, which are permanently stored as a separate dataset. This strategy helps the model learn and enhances its resilience to adversarial attacks. This study also evaluates models by subjecting them to adversarial attacks, such as the One Pixel Attack and the Fast Gradient Sign Method, to identify any potential vulnerabilities. Moreover, we use two different model architectures to see how well they are protected against adversarial attacks. It compared their performances to determine the best model for making systems more resistant while still maintaining good performance. The findings show that the combination of the proposed adversarial training technique and an efficient model architecture outcome in increased resistance to adversarial attacks. This also improves the reliability of the model and saves more resources for computation. This is evidenced by the high accuracy results achieved at 98.81% accuracy on the CK+ datasets. The adversarial training technique proposed in this study offers an efficient alternative to overcome the limitations of computational resources. This fortifies the model against adversarial attacks, resulting in significant increases in model resilience without loss of performance.