Robust Recognition Research Articles

Self- attention based optimized deep convolutional robust character and numeral recognition from Gujarati language

Self- attention based optimized deep convolutional robust character and numeral recognition from Gujarati language

Serotonergic amplification of odor-evoked neural responses maps onto flexible behavioral outcomes.

Behavioral responses to many odorants are not fixed but are flexible, varying based on organismal needs. How such variations arise and the role of various neuromodulators in achieving flexible neural-to-behavioral mapping is not fully understood. In this study, we examined how serotonin modulates the neural and behavioral responses to odorants in locusts (Schistocerca americana). Our results indicated that serotonin can increase or decrease appetitive behavior in an odor-specific manner. On the other hand, in the antennal lobe, serotonergic modulation enhanced odor-evoked response strength but left the temporal features or the combinatorial response profiles unperturbed. This result suggests that serotonin allows for sensitive and robust recognition of odorants. Nevertheless, the uniform neural response amplification appeared to be at odds with the observed stimulus-specific behavioral modulation. We show that a simple linear model with neural ensembles segregated based on behavioral relevance is sufficient to explain the serotonin-mediated flexible mapping between neural and behavioral responses.

Serotonergic amplification of odor-evoked neural responses maps onto flexible behavioral outcomes

Behavioral responses to many odorants are not fixed but are flexible, varying based on organismal needs. How such variations arise and the role of various neuromodulators in achieving flexible neural-to-behavioral mapping is not fully understood. In this study, we examined how serotonin modulates the neural and behavioral responses to odorants in locusts (Schistocerca americana). Our results indicated that serotonin can increase or decrease appetitive behavior in an odor-specific manner. On the other hand, in the antennal lobe, serotonergic modulation enhanced odor-evoked response strength but left the temporal features or the combinatorial response profiles unperturbed. This result suggests that serotonin allows for sensitive and robust recognition of odorants. Nevertheless, the uniform neural response amplification appeared to be at odds with the observed stimulus-specific behavioral modulation. We show that a simple linear model with neural ensembles segregated based on behavioral relevance is sufficient to explain the serotonin-mediated flexible mapping between neural and behavioral responses.

Adaptative machine vision with microsecond-level accurate perception beyond human retina

Visual adaptive devices have potential to simplify circuits and algorithms in machine vision systems to adapt and perceive images with varying brightness levels, which is however limited by sluggish adaptation process. Here, the avalanche tuning as feedforward inhibition in bionic two-dimensional (2D) transistor is proposed for fast and high-frequency visual adaptation behavior with microsecond-level accurate perception, the adaptation speed is over 104 times faster than that of human retina and reported bionic sensors. As light intensity changes, the bionic transistor spontaneously switches between avalanche and photoconductive effect, varying responsivity in both magnitude and sign (from 7.6 × 104 to −1 × 103 A/W), thereby achieving ultra-fast scotopic and photopic adaptation process of 108 and 268 μs, respectively. By further combining convolutional neural networks with avalanche-tuned bionic transistor, an adaptative machine vision is achieved with remarkable microsecond-level rapid adaptation capabilities and robust image recognition with over 98% precision in both dim and bright conditions.

Pyrrole-based acyclic hexapodal aldehyde for anion recognition

A novel acyclic hexapodal host (4a) featuring with multiple formyl groups, pyrrole cores, and ester linkages has been prepared and characterized through 1H NMR, 13C NMR, high-resolution mass spectrometry and single-crystal X-ray diffraction techniques. This versatile host molecule exhibits robust anion recognition for CN−, F−, SO42−, HSO4−, H2PO4−, and HP2O73− with the binding constants spanning 0.18 × 105 M−1 to 7.72 × 105 M−1. Interactions with these anions were elucidated through UV–visible and 1H NMR titrations, complemented by DFT calculations. At low guest concentrations (5- to 100-fold excess), a 1:1 complexation is predominant. However, at high concentrations (1000-fold excess), the host's flexibility promotes the assembly of other host-guest complexes. Comparative studies with related pyrrole-based hexapodal ester 4b and tetrapodal aldehyde 4c highlight the importance of peripheral formyl groups, pyrrolic NH groups, the number of pendant arms, and conformational adaptability in anion recognition. These insights underscore the potential of acyclic multipodal hosts in advancing supramolecular chemistry and anion recognition.

Self-Supervised EEG Representation Learning for Robust Emotion Recognition

Emotion recognition based on electroencephalography (EEG) is becoming a growing concern of researchers due to its various applications and portable devices. Existing methods are mainly dedicated to EEG feature representation and have made impressive progress. However, the problem of scarce labels restricts their further promotion. In light of this, we propose a self-supervised framework with contrastive learning for robust EEG-based emotion recognition, which can effectively leverage both readily available unlabeled EEG signals and labeled ones to learn highly discriminative EEG features. Firstly, we construct a specific pretext task according to the sequential non-stationarity of emotional EEG signals for contrastive learning, which aims at extracting pseudo-label information from all EEG data. Meanwhile, we propose a novel negative segment selection algorithm to reduce the noise of unlabeled data during the contrastive learning process. Secondly, to mitigate the overfitting issue induced by a small number of labeled samples during learning, we originate a loss function with label smoothing regularization that can guide the model to learn generalizable features. Extensive experiments over three benchmark datasets demonstrate the effectiveness and superiority of our model on EEG-based emotion recognition task. Besides, the generalization and robustness of the model have also been proved through sufficient experiments.

Home Activity Recognition for Rural Elderly Based on Deep Learning and Smartphone Sensors

With the exacerbation of the rural aging population trend, home-based health monitoring for the rural elderly has become a societal focal point, demanding an effective technological means to elevate the level of rural elderly health management. However, traditional algorithms for monitoring rural elderly behavior face myriad challenges, such as effectively capturing temporal and spatial features. Consequently, addressing the need to enhance the accuracy and robustness of rural elderly behavior recognition has become an urgent problem to solve. This study responds to this challenge by comprehensively employing deep learning and temporal modeling techniques, designing, and validating a short-term and long-term dual-layer home-based health monitoring system for the rural elderly.In the short-term layer, the model utilizes smartphones to collect health information from the rural elderly in various ways and performs real-time anomaly behavior detection.

Rapid Screening of Emergent Febuxostat Adulteration in Functional Foods Based on a Simulation-Inspired High-Quality Antibody.

Due to the potential health risks of adulterated febuxostat in uric-acid-lowering foods, it is urgent to develop rapid detection methods. However, there are no fast analytical techniques for febuxostat yet. Herein, an efficient hapten simulation strategy was proposed to successfully produce a highly sensitive and selective monoclonal antibody toward febuxostat. Based on such a robust recognition element, easy colorimetric and ultrasensitive fluorescent lateral flow immunochromatographic immunoassays were first established, which can detect febuxostat as low as 60 μg/kg by the naked eye or 1.01 μg/kg by a commercial test strip reader with acceptable stability. Furthermore, in the recovery test and blind sample analysis, consistent results between our methods and the authorized liquid chromatography-tandem mass spectrometry method suggested the high accuracy and practicality of this work. The present work not only proposes a rational hapten design idea but also provides favorable tools for the rapid screening of febuxostat in functional foods.

Trademark Text Recognition Combining SwinTransformer and Feature-Query Mechanisms

The task of trademark text recognition is a fundamental component of scene text recognition (STR), which currently faces a number of challenges, including the presence of unordered, irregular or curved text, as well as text that is distorted or rotated. In applications such as trademark infringement detection and analysis of brand effects, the diversification of artistic fonts in trademarks and the complexity of the product surfaces where the trademarks are located pose major challenges for relevant research. To tackle these issues, this paper proposes a novel recognition framework named SwinCornerTR, which aims to enhance the accuracy and robustness of trademark text recognition. Firstly, a novel feature-extraction network based on SwinTransformer with EFPN (enhanced feature pyramid network) is proposed. By incorporating SwinTransformer as the backbone, efficient capture of global information in trademark images is achieved through the self-attention mechanism and enhanced feature pyramid module, providing more accurate and expressive feature representations for subsequent text extraction. Then, during the encoding stage, a novel feature point-retrieval algorithm based on corner detection is designed. The OTSU-based fast corner detector is presented to generate a corner map, achieving efficient and accurate corner detection. Furthermore, in the encoding phase, a feature point-retrieval mechanism based on corner detection is introduced to achieve priority selection of key-point regions, eliminating character-to-character lines and suppressing background interference. Finally, we conducted extensive experiments on two open-access benchmark datasets, SVT and CUTE80, as well as a self-constructed trademark dataset, to assess the effectiveness of the proposed method. Our results showed that the proposed method achieved accuracies of 92.9%, 92.3% and 84.8%, respectively, on these datasets. These results demonstrate the effectiveness and robustness of the proposed method in the analysis of trademark data.

Perbandingan Dataset Labelled Faces in the Wild (LFW) dan faces94 Menggunakan Algoritma Convolutional Neural Networks (CNN) untuk Pengenalan Wajah

This research compares the performance of two popular datasets, Labelled Faces in the Wild (LFW) and faces94, in the task of face recognition using Convolutional Neural Networks (CNN) algorithms. The LFW dataset is known for its high variation in pose, lighting, and expression, while faces94 is more structured with more uniform lighting and pose conditions. CNNs were chosen for their ability to extract important features from face images for classification. In this study, a CNN model was trained on both datasets and its performance was evaluated using accuracy, precision, and recall metrics. The experimental results showed that the model trained on the faces94 dataset achieved higher accuracy compared to the model trained on the LFW dataset. However, the model on the LFW dataset demonstrated better resilience to variations in lighting and pose conditions. These findings indicate that while a more structured dataset like faces94 can produce a model with high accuracy under testing conditions similar to the training data, a dataset with greater variation like LFW is more suitable for real-world applications involving diverse conditions. This study provides important insights into the selection of datasets for developing robust and accurate face recognition systems.

Gesture-Controlled Robotic Arm for Agricultural Harvesting Using a Data Glove with Bending Sensor and OptiTrack Systems.

This paper presents a gesture-controlled robotic arm system designed for agricultural harvesting, utilizing a data glove equipped with bending sensors and OptiTrack systems. The system aims to address the challenges of labor-intensive fruit harvesting by providing a user-friendly and efficient solution. The data glove captures hand gestures and movements using bending sensors and reflective markers, while the OptiTrack system ensures high-precision spatial tracking. Machine learning algorithms, specifically a CNN+BiLSTM model, are employed to accurately recognize hand gestures and control the robotic arm. Experimental results demonstrate the system's high precision in replicating hand movements, with a Euclidean Distance of 0.0131 m and a Root Mean Square Error (RMSE) of 0.0095 m, in addition to robust gesture recognition accuracy, with an overall accuracy of 96.43%. This hybrid approach combines the adaptability and speed of semi-automated systems with the precision and usability of fully automated systems, offering a promising solution for sustainable and labor-efficient agricultural practices.

Efficient YOLOv8 algorithm for extreme small-scale object detection

The past decade has witnessed significant progress in detecting small objects that are often distributed with large-scale variations in arbitrary orientations. However, most of the existing models are unable to detect extremely small objects, due to (1) tiny-scale objects with more blurring in comparison to ground-view objects provide less important information for precise and robust recognition; (2) Unevenly distributed objects make detection inefficient, particularly in regions occupied by dense objects, which lead to the common inconsistency between the classification score and localization accuracy. This research proposes an efficient YOLOv8 model with a new backbone network, and anchor-free head to address these issues. Bottleneck layers are used in the backbone, allowing the model to capture more complicated patterns and reduce the dimensionality of the feature maps while maintaining the same number of channels. Whereas an anchor-free head detection module does not require pre-defined anchor boxes, they directly predict objects’ center points and sizes and their corresponding class probabilities. This proposed model is experimentally verified on the VisDrone dataset and achieved higher mAP (45.9 %), and precision (76.8 %). These experimental results suggest that the proposed model outperforms existing models in detecting x-small items.

Enhancing autonomous driving safety: A robust traffic sign detection and recognition model TSD-YOLO

As autonomous driving technology rapidly advances, Traffic Sign Detection and Recognition (TSDR) has become pivotal in ensuring the safety and regulatory compliance of autonomous vehicles. Despite progress, existing technologies struggle under challenging conditions such as adverse weather and complex roadway environments. To overcome these obstacles, we introduce a novel model, TSD-YOLO, which leverages Mamba and YOLO technologies to enhance the accuracy and robustness of traffic sign detection. Our innovative YOLO-MAM dual-branch module merges convolutional layer-based local feature extraction with the long-distance dependency capabilities of the State Space Models (SSMs). We conducted experimental validations using the Tsinghua-Tencent 100K (TT-100K) dataset and the Mapillary Traffic Sign Detection (MTSD) dataset, demonstrating our model’s efficacy across various datasets. Furthermore, cross-dataset validations affirm the model’s exceptional generalization and robustness across diverse environments. This study not only bolsters traffic sign detection and recognition in autonomous driving systems but also paves the way for future advancements in autonomous driving technology.

TRNet: Two-level Refinement Network leveraging speech enhancement for noise robust speech emotion recognition

One persistent challenge in Speech Emotion Recognition (SER) is the ubiquitous environmental noise, which frequently results in deteriorating SER performance in practice. In this paper, we introduce a Two-level Refinement Network, dubbed TRNet, to address this challenge. Specifically, a pre-trained speech enhancement module is employed for front-end noise reduction and noise level estimation. Later, we utilize clean speech spectrograms and their corresponding deep representations as reference signals to refine the spectrogram distortion and representation shift of enhanced speech during model training. Experimental results validate that the proposed TRNet substantially promotes the robustness of the proposed system in both matched and unmatched noisy environments, without compromising its performance in noise-free environments.

Robust gesture recognition based on attention-deep fast convolutional neural network and surface electromyographic signals.

The surface electromyographic (sEMG) signals reflect human motor intention and can be utilized for human-machine interfaces (HMI). Comparing to the sparse multi-channel (SMC) electrodes, the high-density (HD) electrodes have a large number of electrodes and compact space between electrodes, which can achieve more sEMG information and have the potential to achieve higher performance in myocontrol. However, when the HD electrodes grid shift or damage, it will affect gesture recognition and reduce recognition accuracy. To minimize the impact resulting from the electrodes shift and damage, we proposed an attention deep fast convolutional neural network (attention-DFCNN) model by utilizing the temporary and spatial characteristics of high-density surface electromyography (HD-sEMG) signals. Contrary to the previous methods, which are mostly base on sEMG temporal features, the attention-DFCNN model can improve the robustness and stability by combining the spatial and temporary features. The performance of the proposed model was compared with other classical method and deep learning methods. We used the dataset provided by The University Medical Center Göttingen. Seven able-bodied subjects and one amputee involved in this work. Each subject executed nine gestures under the electrodes shift (10 mm) and damage (6 channels). As for the electrodes shift 10 mm in four directions (inwards; onwards; upwards; downwards) on seven able-bodied subjects, without any pre-training, the average accuracy of attention-DFCNN (0.942 0.04) is significantly higher than LSDA (0.910 0.04, p < 0.01), CNN (0.920 0.05, p < 0.01), TCN (0.840 0.07, p < 0.01), LSTM (0.864 0.08, p < 0.01), attention-BiLSTM (0.852 0.07, p < 0.01), Transformer (0.903 0.07, p < 0.01) and Swin-Transformer (0.908 0.09, p < 0.01). The proposed attention-DFCNN algorithm and the way of combining the spatial and temporary features of sEMG signals can significantly improve the recognition rate when the HD electrodes grid shift or damage during wear.

A survey for table recognition based on deep learning

Tables, as an important information interaction media containing much structured information, are popularly adopted in our life. There are a lot of table recognition methods for understanding and applying these information. Among them, table recognition based on deep learning is the mainstream. Therefore, we conducted this survey to summarize the latest research on table recognition based on deep learning in recent years. Table recognition can be divided into three sub-tasks, namely table detection, table structure recognition and table content recognition, so we firstly introduce the importance and significance of table recognition research and analyze the current challenges faced by each sub-task. Then, we list the existing public table recognition datasets and introduce commonly used evaluation metrics. Furthermore, we review deep learning-based solutions proposed in recent years for three different sub-tasks and reclassify them. Especially for table structure recognition, we reclassify them according to their annotation schemes. Finally, we report the performance of some excellent methods, such as CascadeTabNet, TSRFormer, SEM, etc, on public datasets, and describe the future research directions, including table recognition in natural scenes, question and answer about tables, visual information extraction, domain adaptation etc. In summary, this survey reviews the data, evaluations, methods, and development directions of table recognition. Table understanding and domain adaptation need as well to be focused on. The former can predict the relationships between two table elements, which helps us to understand the table. The latter can enhance the generalization ability of table recognition methods, thereby obtaining more robust recognition performance. Both are important development directions for table recognition in the future.

Advancing robust underwater acoustic target recognition through multitask learning and multi-gate mixture of experts.

Underwater acoustic target recognition has emerged as a prominent research area within the field of underwater acoustics. However, the current availability of authentic underwater acoustic signal recordings remains limited, which hinders data-driven acoustic recognition models from learning robust patterns of targets from a limited set of intricate underwater signals, thereby compromising their stability in practical applications. To overcome these limitations, this study proposes a recognition framework called M3 (multitask, multi-gate, multi-expert) to enhance the model's ability to capture robust patterns by making it aware of the inherent properties of targets. In this framework, an auxiliary task that focuses on target properties, such as estimating target size, is designed. The auxiliary task then shares parameters with the recognition task to realize multitask learning. This paradigm allows the model to concentrate on shared information across tasks and identify robust patterns of targets in a regularized manner, thus, enhancing the model's generalization ability. Moreover, M3 incorporates multi-expert and multi-gate mechanisms, allowing for the allocation of distinct parameter spaces to various underwater signals. This enables the model to process intricate signal patterns in a fine-grained and differentiated manner. To evaluate the effectiveness of M3, extensive experiments were implemented on the ShipsEar underwater ship-radiated noise dataset. The results substantiate that M3 has the ability to outperform the most advanced single-task recognition models, thereby achieving the state-of-the-art performance.

Multi-scale keypoints detection and motion features extraction in dairy cows using ResNet101-ASPP network

Detecting Keypoints in dairy cows aims to locate and track the motion trajectories of the body's joints, which plays a crucial role in behavior analysis and lameness detection. However, real farming scenarios, characterized by occlusions and large variations in object scale may result in poor detection results. Therefore we introduce the atrous spatial pyramid pooling (ASPP) module into the shallow network of ResNet101, designed to improve the multi-scale feature extraction capability of the model. The ASPP module enhances the robustness of recognition for different dimensional sizes and occluded keypoints using different dilatation rates in the parallel atrous convolutional layers to expand the model's receptive field. Furthermore, seven types of motion features, including tracking up, gait symmetry, step height balance, motion speed variability, head swing amplitude, head-neck slope and back curvature are extracted simultaneously by monitoring and tracking the motion trajectory of distinct keypoints. Several of these features represent innovative extraction models and attributes, first proposed in this study. Multiple models are trained and tested on datasets containing 2385 frames for ablation experiments. The experiments show that, in comparison with the ResNet50, MobileNet_v2_1.0, and EfficientNet-b0 backbone networks, the training error and test error of ResNet101 improve by 4.04–30.12 pixels and 3.81–28.14 pixels. Therefore, ResNet101 is used as the benchmark for subsequent model improvement by adding the ASPP module. The training error and test error of the ResNet101-ASPP network are improved by 0.27 and 0.24 pixels, respectively, compared to the benchmark network. The prediction confidence improves by 1.65-2.50% at three different dairy cow object scales, In addition, the keypoints under different occlusion conditions improve considerably, especially for small-scale keypoints, demonstrating the capability of the ASPP module for multi-scale feature extraction. By analyzing the distribution between the seven features and health, mild lameness, and severe lameness in dairy cows, it is shown that all the different features play an important role in distinguishing between different levels of lameness.

Development of a Face Recognition System for Authentication in Computer-Based Examinations: A Comparison with Fingerprint Verification

This study aims to overcome the limitations of fingerprint biometric authentication in computer-based examinations by introducing a robust face recognition system. The research targets key challenges such as administrative overhead, prolonged authentication times, and issues with false positives and negatives. By leveraging advanced technologies like Multi-Task Cascaded Convolutional Neural Network (MTCNN) for precise facial detection and FaceNet for efficient facial feature embedding, the developed system achieves substantial improvements across critical metrics. The evaluation metrics consistently report high ratings: accuracy 90%, speed 92%, ease of use 86%, robustness to variations 94%, security 92%, and privacy 90%. Evaluation of the system shows its superiority over traditional fingerprint methods. It demonstrates notably higher accuracy, efficiency, user-friendliness, adaptability to diverse conditions, and enhanced levels of security and privacy. These advancements position the face recognition system as a compelling alternative for enhancing the authentication processes in computer-based examinations. Keywords: FaceNet, Multi-Task Cascaded Convolutional Neural Network, Face Recognition, Security, Face, Fingerprint Detection. Sulaimon O.O., Olabiyisi, S.O. &amp; Ismaila, W.O. (2024): Development of a Face Recognition System for Authentication in Computer-Based Examinations: A Comparison with Fingerprint Verification. Journal of Advances in Mathematical &amp; Computational Science. Vol. 12, No. 2. Pp 117-128. Available online at www.isteams.net/mathematics-computationaljournal. dx.doi.org/10.22624/AIMS/MATHS/V12N2P8

Collaborative Analysis of Learners’ Emotional States Based on Cross-Modal Higher-Order Reasoning

Emotion is a significant factor influencing education and teaching, closely intertwined with learners’ cognitive processing. Conducting analysis of learners’ emotions based on cross-modal data is beneficial for achieving personalized guidance in intelligent educational environments. Currently, due to factors such as data scarcity and environmental noise, data imbalances have led to incomplete or missing emotional information. Therefore, this study proposes a collaborative analysis model based on attention mechanisms. The model extracts features from various types of data using different tools and employs multi-head attention mechanisms for parallel processing of feature vectors. Subsequently, through a cross-modal attention collaborative interaction module, effective interaction among visual, auditory, and textual information is facilitated, significantly enhancing comprehensive understanding and the analytical capabilities of cross-modal data. Finally, empirical evidence demonstrates that the model can effectively improve the accuracy and robustness of emotion recognition in cross-modal data.