Existing Recognition Models Research Articles

Transfer-learning enabled micro-expression recognition using dense connections and mixed attention

Micro-expression recognition (MER) is a challenging computer vision problem, where the limited amount of available training data and insufficient intensity of the facial expressions are among the main issues adversely affecting the performance of existing recognition models. To address these challenges, this paper explores a transfer–learning enabled MER model using a densely connected feature extraction module with mixed attention. Unlike previous works that utilize transfer learning to facilitate MER and extract local facial-expression information, our model relies on pretraining with three diverse macro-expression datasets and, as a result, can: (i) overcome the problem of insufficient sample size and limited training data availability, (ii) leverage (related) domain-specific information from multiple datasets with diverse characteristics, and (iii) improve the model adaptability to complex scenes. Furthermore, to enhance the intensity of the micro-expressions and improve the discriminability of the extracted features, the Euler video magnification (EVM) method is adopted in the preprocessing stage and then used jointly with a densely connected feature extraction module and a mixed attention mechanism to derive expressive feature representations for the classification procedure. The proposed feature extraction mechanism not only guarantees the integrity of the extracted features but also efficiently captures local texture cues by aggregating the most salient information from the generated feature maps, which is key for the MER task. The experimental results on multiple datasets demonstrate the robustness and effectiveness of our model compared to the state-of-the-art.

A Deformable and Multi-Scale Network with Self-Attentive Feature Fusion for SAR Ship Classification

The identification of ships in Synthetic Aperture Radar (SAR) imagery is critical for effective maritime surveillance. The advent of deep learning has significantly improved the accuracy of SAR ship classification and recognition. However, distinguishing features between different ship categories in SAR images remains a challenge, particularly as the number of categories increases. The key to achieving high recognition accuracy lies in effectively extracting and utilizing discriminative features. To address this, we propose DCN-MSFF-TR, a novel recognition model inspired by the Transformer encoder–decoder architecture. Our approach integrates a deformable convolutional module (DCN) within the backbone network to enhance feature extraction. Additionally, we introduce multi-scale self-attention processing from the Transformer into the feature hierarchy and fuse these representations at appropriate levels using a feature pyramid strategy. This enables each layer to leverage both its own information and synthesized features from other layers, enhancing feature representation. Extensive evaluations on the OpenSARShip-3-Complex and OpenSARShip-6-Complex datasets demonstrate the effectiveness of our method. DCN-MSFF-TR achieves average recognition accuracies of 78.1% and 66.7% on the three-class and six-class datasets, respectively, outperforming existing recognition models and showcasing its superior capability in accurately identifying ship categories in SAR images.

A Lightweight Rice Pest Detection Algorithm Using Improved Attention Mechanism and YOLOv8

Intelligent pest detection algorithms are capable of effectively detecting and recognizing agricultural pests, providing important recommendations for field pest control. However, existing recognition models have shortcomings such as poor accuracy or a large number of parameters. Therefore, this study proposes a lightweight and accurate rice pest detection algorithm based on improved YOLOv8. Firstly, a Multi-branch Convolutional Block Attention Module (M-CBAM) is constructed in the YOLOv8 network to enhance the feature extraction capability for pest targets, yielding better detection results. Secondly, the Minimum Points Distance Intersection over Union (MPDIoU) is introduced as a bounding box loss metric, enabling faster model convergence and improved detection results. Lastly, lightweight Ghost convolutional modules are utilized to significantly reduce model parameters while maintaining optimal detection performance. The experimental results demonstrate that the proposed method outperforms other detection models, with improvements observed in all evaluation metrics compared to the baseline model. On the test set, this method achieves a detection average precision of 95.8% and an F1-score of 94.6%, with a model parameter of 2.15 M, meeting the requirements of both accuracy and lightweightness. The efficacy of this approach is validated by the experimental findings, which provide specific solutions and technical references for intelligent pest detection.

LCA-Net: A Lightweight Cross-Stage Aggregated Neural Network for Fine-Grained Recognition of Crop Pests and Diseases

In the realm of smart agriculture technology’s rapid advancement, the integration of various sensors and Internet of Things (IoT) devices has become prevalent in the agricultural sector. Within this context, the precise identification of pests and diseases using unmanned robotic systems assumes a crucial role in ensuring food security, advancing agricultural production, and maintaining food reserves. Nevertheless, existing recognition models encounter inherent limitations such as suboptimal accuracy and excessive computational efforts when dealing with similar pests and diseases in real agricultural scenarios. Consequently, this research introduces the lightweight cross-layer aggregation neural network (LCA-Net). To address the intricate challenge of fine-grained pest identification in agricultural environments, our approach initially enhances the high-performance large-scale network through lightweight adaptation, concurrently incorporating a channel space attention mechanism. This enhancement culminates in the development of a cross-layer feature aggregation (CFA) module, meticulously engineered for seamless mobile deployment while upholding performance integrity. Furthermore, we devised the Cut-Max module, which optimizes the accuracy of crop pest and disease recognition via maximum response region pruning. Thorough experimentation on comprehensive pests and disease datasets substantiated the exceptional fine-grained performance of LCA-Net, achieving an impressive accuracy rate of 83.8%. Additional ablation experiments validated the proposed approach, showcasing a harmonious balance between performance and model parameters, rendering it suitable for practical applications in smart agricultural supervision.

InOR-Net: Incremental 3-D Object Recognition Network for Point Cloud Representation.

3-D object recognition has successfully become an appealing research topic in the real world. However, most existing recognition models unreasonably assume that the categories of 3-D objects cannot change over time in the real world. This unrealistic assumption may result in significant performance degradation for them to learn new classes of 3-D objects consecutively due to the catastrophic forgetting on old learned classes. Moreover, they cannot explore which 3-D geometric characteristics are essential to alleviate the catastrophic forgetting on old classes of 3-D objects. To tackle the above challenges, we develop a novel Incremental 3-D Object Recognition Network (i.e., InOR-Net), which could recognize new classes of 3-D objects continuously by overcoming the catastrophic forgetting on old classes. Specifically, category-guided geometric reasoning is proposed to reason local geometric structures with distinctive 3-D characteristics of each class by leveraging intrinsic category information. We then propose a novel critic-induced geometric attention mechanism to distinguish which 3-D geometric characteristics within each class are beneficial to overcome the catastrophic forgetting on old classes of 3-D objects while preventing the negative influence of useless 3-D characteristics. In addition, a dual adaptive fairness compensations' strategy is designed to overcome the forgetting brought by class imbalance by compensating biased weights and predictions of the classifier. Comparison experiments verify the state-of-the-art performance of the proposed InOR-Net model on several public point cloud datasets.

A Knowledge Inference and Sharing-Based Open-Set Device Recognition Approach for Satellite-Terrestrial-Integrated IoT

Satellite-terrestrial-integrated internet of things (IoT) is an inevitable trend in future development, but open satellite link and massive IoT device access will bring serious security risks. However, most existing recognition models are unable to discover and reject malicious IoT devices since they lack the decision information of these unauthorized devices during training. To address this dilemma, this paper proposes a knowledge inference and sharing-based open-set recognition approach to protect satellite-terrestrial-integrated IoT. It proceeds in two steps. First, knowledge inference, where we construct ideal substitutes for unauthorized devices after reasonable inference on the training set, aims to compensate the model’s missing decision information. Second, knowledge sharing, where we inherit the existing knowledge and modify the model’s decision boundaries through model expansion and knowledge distillation, achieves accurate open-set recognition. Experiments on the ORACLE dataset demonstrated that our approach outperforms other state-of-the-art OSR methods in terms of accuracy and running time. In short, our approach has excellent performance while only slightly increasing computational complexity.

A hybrid location‐dependent ultra convolutional neural network‐based vehicle number plate recognition approach for intelligent transportation systems

SummaryIn today's world, identifying the owner and proprietor of a vehicle that violates driving rules or does any unintentional work on the street is a challenging task. Inspection of each driver's license number takes a long time for a highway police officer. To overcome this, many researchers have introduced an automated number plate recognition approach which is usually a computer vision‐based technique to identify the vehicle's registration plate. However, the existing recognition approaches are lagged to extract the influential features which degrade the detection accuracy and increase the misclassification errors. In this article, a novel automated number plate recognition methodology has been proposed to identify the number plates accurately with minimal error rates. Primary, a new pretrained location‐dependent ultra convolutional neural network (LUCNN) is employed to learn the influential features from the input images. These obtained features are then fed into hybrid single‐shot fully convolutional detectors with a support vector machine (SSVM) classifier to separate the vehicle's city, model, and number from the registration location. At varied automobile distances, the proposed LUCNN + SSVM model is able to retrieve the number plate regions in the picture acquired from its back end. The performance results manifest that the proposed LUCNN + SSVM model attains a better accuracy of 98.75% and a lesser error range of 1.25% than the existing recognition models.

When Mobilenetv2 Meets Transformer: A Balanced Sheep Face Recognition Model

Sheep face recognition models deployed on edge devices require a good trade-off between model size and accuracy, but the existing recognition models cannot do so. To solve the above problems, this paper combines Mobilenetv2 with Vision Transformer to propose a balanced sheep face recognition model called MobileViTFace. MobileViTFace enhances the model’s ability to extract fine-grained features and suppress the interference of background information through Transformer to distinguish different sheep faces more effectively. Thus, it can distinguish different sheep faces more effectively. The recognition accuracy of 96.94% is obtained on a self-built dataset containing 5490 sheep face photos of 105 sheep, which is a 9.79% improvement compared with MobilenetV2, with only a small increase in Params (the number of parameters) and FLOPs (floating-point operations). Compared to models such as Swin-small, which currently performs SOTA, Params and FLOPs are reduced by nearly ten times, whereas recognition accuracy is only 0.64% lower. Deploying MobileViTFace on the Jetson Nano-based edge computing platform, real-time and accurate recognition results are obtained, which has implications for practical production.

Improvement of Identity Recognition with Occlusion Detection-Based Feature Selection

Image-based facial identity recognition has become a technology that is now used in many applications. This is because it is possible to use only a camera without the need for any other device. Besides, due to the advantage of contactless technology, it is one of the most popular certifications. However, a common recognition system is not possible if some of the face information is lost due to the user’s posture or the wearing of masks, as caused by the recent prevalent disease. In some platforms, although performance is improved through incremental updates, it is still inconvenient and inaccurate. In this paper, we propose a method to respond more actively to these situations. First, we determine whether an obscurity occurs and improve the stability by calculating the feature vector using only a significant area when the obscurity occurs. By recycling the existing recognition model, without incurring little additional costs, the results of reducing the recognition performance drop in certain situations were confirmed. Using this technique, we confirmed a performance improvement of about 1~3% in a situation where some information is lost. Although the performance is not dramatically improved, it has the big advantage that it can improve recognition performance by utilizing existing systems.

Sintering conditions recognition of rotary kiln based on kernel modification considering class imbalance

Accurate sintering condition recognition (SCR) is an important precondition for optimal control of rotary kilns. However, the occurrence probability of abnormal conditions in the industrial field is much lower than normal, resulting in imbalanced class sintering samples in general. This significantly deteriorates the effectiveness of existing recognition models in abnormal condition detection. In this paper, an integrated framework considering class imbalance is proposed for sintering condition recognition. In the proposed framework, after analysing the characteristics of thermal signals by the Lipschitz method, four discriminant features are extracted to comprehensively describe different sintering conditions. In addition, focusing on the class imbalance of sintering samples, the kernel modification method is introduced to enhance the optimal marginal distribution machine (ODM), and a novel recognition model kernel modified the ODM (KMODM) is proposed for SCR. By constructing a new conformal transformation function to modify the ODM kernel function, KMODM optimizes the spatial distribution of training samples in the kernel space, thereby alleviating the detection accuracy deterioration of the minority class. The experimental results on real thermal signals and standard datasets show that the KMODM model can effectively handle imbalanced data. Based on this, the proposed SCR framework can reduce the misjudgement of abnormal conditions and balance the recognition accuracy of each condition.

Multi-Pose Face Recognition Based on Deep Learning in Unconstrained Scene

At present, deep learning drives the rapid development of face recognition. However, in the unconstrained scenario, the change of facial posture has a great impact on face recognition. Moreover, the current model still has some shortcomings in accuracy and robustness. The existing research has formulated two methods to solve the above problems. One method is to model and train each pose separately. Then, a fusion decision will be made. The other method is to make “frontal” faces on the image or feature level and transform them into “frontal” face recognition. Based on the second idea, we propose a profile to the frontal revise mapping (PTFRM) module. This module realizes the revision of arbitrary poses on the feature level and transforms the multi-pose features into an approximate frontal representation to enhance the recognition ability of the existing recognition models. Finally, we evaluate the PTFRM on unconstrained face validation benchmark datasets such as Labeled Faces in the Wild (LFW), Celebrities in Frontal Profile (CFP), and IARPA Janus Benchmark A(IJB-A). Results show that the chosen method for this study achieves good performance.

Optimizing Features Quality: A Normalized Covariance Fusion Framework for Skeleton Action Recognition

Action recognition based on 3D skeleton sequences has gained considerable attention in recent years. Due to effectively representing the spatial and the temporal characters of skeleton sequences, the Covariance Matrix (CM) features combined with the Long Short-Term Memory (LSTM) network is an effective and reasonable roadmap to enhance the action recognition accuracy. However, the CM features in the existing recognition models are computed from the raw data without normalization or with static normalization. Moreover, a CM feature is calculated from all coordinates in one frame, treating all coordinates in three axes identically and neglecting the relationship of the coordinates in the same axe. In this paper, an end to end deep learning framework is proposed that includes a normalization layer dynamically adapting to data distribution and inference procedure. After normalization, the three covariance feature sequences from the coordinates in three axes are produced from the sliding windows and are fused into one fusion matrix using a convolution layer. Finally, the fusion matrix is sequentially fed into an LSTM network to recognize skeleton action. The novelty of the proposed framework is combining the adaptive preprocessing and the features fusion to the LSTM network and improving the recognition accuracy by optimizing the quality of the features rather than network construction. In the experiments, the proposed framework is verified on the public datasets and one student action dataset collected from a real classroom. The experimental results demonstrate that the proposed method achieves a significant improvement in accuracy compared to the state-of-the-art methods. It can be concluded that the proposed framework can not only accurately capture the correlation of joints in the same frame but can also effectively express the dependences of sequential frames.

Usefulness Individualized Recognition Model of Consumer Online Reviews Based on Transfer Learning

Usefulness recognition model of consumer online reviews can filter out useless ones in all reviews, which can reduce cost of searching for commodity information for potential consumers. However, existing recognition models were based on that all consumers judge usefulness of online reviews by same criteria, ignoring differences of theirs. Thus, a new transfer learning support vector machine model was proposed in this paper to realize individualized recognition of online reviews usefulness. Theoretical analysis and experimental results showed that proposed algorithm could achieve better performance with less training data.

Accounting for the build-up of proactive interference across lists in a list length paradigm reveals a dominance of item-noise in recognition memory

Abstract There has been a longstanding debate concerning whether interference in recognition memory is attributable to other items on the study list (i.e., item-noise) or to prior memories (i.e., context-noise and background-noise). Recently, Osth and Dennis (2015) devised a global matching model that could estimate the magnitude of each interference contribution and they found that context-noise and background-noise were dominant in recognition. In the present investigation, data from a list length experiment were analysed using variants of the Osth, Jansson, Dennis, and Heathcote (2018) model, that integrates the memory retrieval components of the Osth and Dennis (2015) model with the diffusion decision model (Ratcliff, 1978) to jointly account for choice probabilities and RT distributions. The standard version of the model, like existing recognition models, treated each condition as if no proactive interference had accumulated over the session. A more comprehensive version of the model allowed both study and test items from prior conditions to contribute proactive interference (PI) to future conditions. While the standard model estimated a dominance of background-noise, the PI model estimated a dominance of item-noise, reversing the conclusions made by Osth and Dennis (2015). Along with list length, the experimental design provided a measure of the test position effect (TPE). While the standard model attributed the TPE to context drift, the PI model attributed the TPE to both context drift and increases in item-noise.

A Novel Fiber Intrusion Signal Recognition Method for OFPS Based on SCN With Dropout

Accurate and fast recognition of fiber intrusion signals has always been a fundamental task in the Optical Fiber Pre-warning System (OFPS). However, currently existing recognition models tend to focus on one aspect and lack a comprehensive approach. In this paper, a dropout-based Stochastic Configuration Network (SCN) optical fiber intrusion signal recognition model is first proposed, which is named DropoutSCN. By combining dropout with randomized algorithm models, it not only enhances the fast learning ability, but also improves the generalization performance of the recognition model. In the experiment, compared with traditional Artificial Neural Network (ANN), Random Vector Functional Link (RVFL), and original SCN models, the DropoutSCN model proposed in this paper has the lowest root-mean-square error (RMSE). In terms of time efficiency, it reduces the time delay by about 2.5 times compared with the traditional ANN. In addition, this paper applies dropout to SCN, which provides a feasible thinking and reference for the study of randomized algorithm models.

Large-Scale Aerial Image Categorization Using a Multitask Topological Codebook.

Fast and accurately categorizing the millions of aerial images on Google Maps is a useful technique in pattern recognition. Existing methods cannot handle this task successfully due to two reasons: 1) the aerial images' topologies are the key feature to distinguish their categories, but they cannot be effectively encoded by a conventional visual codebook and 2) it is challenging to build a realtime image categorization system, as some geo-aware Apps update over 20 aerial images per second. To solve these problems, we propose an efficient aerial image categorization algorithm. It focuses on learning a discriminative topological codebook of aerial images under a multitask learning framework. The pipeline can be summarized as follows. We first construct a region adjacency graph (RAG) that describes the topology of each aerial image. Naturally, aerial image categorization can be formulated as RAG-to-RAG matching. According to graph theory, RAG-to-RAG matching is conducted by enumeratively comparing all their respective graphlets (i.e., small subgraphs). To alleviate the high time consumption, we propose to learn a codebook containing topologies jointly discriminative to multiple categories. The learned topological codebook guides the extraction of the discriminative graphlets. Finally, these graphlets are integrated into an AdaBoost model for predicting aerial image categories. Experimental results show that our approach is competitive to several existing recognition models. Furthermore, over 24 aerial images are processed per second, demonstrating that our approach is ready for real-world applications.

Detection of Financial Reporting Fraud Based on Clustering Algorithm of Automatic Gained Parameter K Value

Listed companies' financial reporting fraud has been a major problem in the research history of accounting. It produces an extremely bad and wide range of influence on the development of securities market. With the continuous development and progress of the stock market, the requirements for strictly controlling and preventing financial reporting fraud are also increasingly high. There have been a lot of studies of financial fraud at home and abroad. They are usually about motivations, means, identification and controlling of financial fraud. Financial fraud recognition is usually divided into signal judgment and model identification. However, the existing recognition models’ accuracy is generally not high. There is a large room for improvement and the models are not applicable enough. In addition, in the era of knowledge economy, with the continuous development of information networks, computer network technology is more and more generally applied in the field of finance. Especially the use of computers in financial reporting fraud investigation can greatly reduce the manpower and resources as well as improve the efficiency of identification. But it is not known that what kind of method combined with computer technology can better identify financial reporting fraud. In this case, the paper aims at establishing an accurate financial reporting fraud recognition model based clustering method.

1A1-L10 オンライン学習型マルコフネットワークに基づく机上動作の識別法(人間機械協調)

Human motion has individual differences in motion trajectories and velocity of parts of the body by age, sex, and physical frame. This sometimes declines action recognition performance when the target subject is changed. In this paper, we propose a method for constructing the action recognition model that fits each person's motion. In the proposed framework, the existing recognition model for some person is modified to fit new person's motion. Only when the label predicted by the existing model is wrong, the model is modified with corrected labels. So, little training data are needed and the recognition model can be constructed more effectively. Empirical result using labeled motion data shows the effectiveness of the proposed framework.

Recognizing spatial patterns: a noisy exemplar approach

Models of categorization typically rely on the use of stimuli composed of well-defined dimensions (e.g., Ashby & Maddox (1998) in Choice, decision, and measurement: Essays in honor of R. Duncan Luce, p. 251–301, Mahwah, NJ: Erlbaum). We apply a similar approach to the analysis of recognition memory. Using a version of short-term recognition paradigm (Sternberg, Science 153 (1966) 652), we asked whether NEMO Sternberg's, a noisy exemplar summed-similarity model, could account for variation in mean performance on individual trials. NEMO provided a very good overall fit to recognition data from three experiments. However, its failure to fit data for certain lists of stimuli suggested a revision of the summed-similarity assumption. Our model-based analysis showed that subjects used interitem similarity, in addition to probe-item similarity, as the basis for their decisions. This represents a major departure from existing recognition models that assume subjects' judgments depend exclusively on the summed similarity of the probe to the study items.