International Journal of Pattern Recognition and Artificial Intelligence

In the field of human–computer interaction, there is a trend toward using emotional interaction and putting emotions at the center. However, traditional facial expression recognition is greatly affected by lighting, skin color and make-up, and is not well recognized. This paper addresses these shortcomings in the following areas: (1) The algorithm of infrared face feature localization is studied, using a combination of region growing algorithm to highlight the organ features of infrared faces, and the Harris corner point detection algorithm to detect the organ features of human beings and locate the location of the large organs related to expressions, such as eyes, nose and mouth. (2) The algorithm for feature extraction is investigated, and the local binarization LBP algorithm is proposed to extract the texture features of the infrared face. The LBP algorithm is synthesized to extract the overall LBP features by applying the innovation of chunking and layering of features. By analyzing and processing the infrared face image data from both visible and infrared face expression data, feature detection, feature extraction and effective classification of the four basic face expressions — happy, angry, disgusted and sad — are performed on the infrared images. The application of the algorithm is processed and improved accordingly to the characteristics of infrared face images, which improves the accuracy of recognition and classification compared to traditional methods.

Convolutional neural networks (CNNs) have been enormously successful in a variety of image recognition tasks. Robustness is an important metric to evaluate the quality of CNNs. However, recent research shows that CNNs are particularly vulnerable to adversarial attacks. This paper proposes an adversarial defense method to increase the robustness of CNNs, namely, SCADefender. The proposed method trains a reformer on adversarial examples and the training set of a target classifier. The architecture of the reformer is stacked convolutional autoencoder. The adversarial examples are generated by using various adversarial attacks such as untargeted FGSM, untargeted CW $L_2$ and untargeted BIS. Given an input image, the trained reformer could remove the adversarial perturbations with a low computational cost. To demonstrate the effectiveness, the proposed method is compared with PuVAE, MagNet, and adversarial training on three well-known datasets including MNIST, Fashion-MNIST, and CIFAR-10. In terms of the average detection rate, the proposed method outperforms other methods. While the proposed method achieves an average detection rate of 97.78% for MNIST, 90.43% for Fashion-MNIST, and 80.64% for CIFAR-10, the comparable methods achieve only 23.69- 86.18% for MNIST, 63.90-79.70% for Fashion-MNIST, and 25.55-77.36% for CIFAR-10.

Multi-contextualized representations learning is vital for named entity recognition (NER), which is a fundamental task for effectively extracting structured information from unstructured text, and forming knowledge bases. This task is particularly challenging when dealing with Chinese text given the absence of evident word boundaries. Chinese word segmentation (CWS) can be leveraged to recognize word boundaries, but named entities often encompass multiple segmented words, making it crucial to use boundary information to correctly recognize and distinguish the relationships between these words. In this paper, we propose MCA-NER, a multi-contextualized adversarial-based attentional deep learning approach for Chinese NER, which combines CWS and part-of-speech (POS) tagging information with the classic BiLSTM-CRF NER model, using adversarial multi-task learning. The model incorporates several self-attention components for adversarial and multi-task learning, effectively synthesizing task-specific and common information attribution while improving performance across all three tasks. Experimental results on the three datasets provide compelling evidence that supports the effectiveness and performance of our model.

Graph neural networks (GNNs) have achieved impressive success in various applications. However, training dedicated GNNs for small-scale graphs still faces many problems such as over-fitting and deficiencies in performance improvements. Traditional methods such as data augmentation are commonly used in computer vision (CV) but are barely applied to graph structure data to solve these problems. In this paper, we propose a training framework named MTDA (Multi-Task learning with Data Augmentation)-GNN, which combines data augmentation and multi-task learning to improve the node classification performance of GNN on small-scale graph data. First, we use Graph Auto-Encoders (GAE) as a link predictor, modifying the original graphs’ topological structure by promoting intra-class edges and demoting interclass edges, in this way to denoise the original graph and realize data augmentation. Then the modified graph is used as the input of the node classification model. Besides defining the node pair classification as an auxiliary task, we introduce multi-task learning during the training process, forcing the predicted labels to conform to the observed pairwise relationships and improving the model’s classification ability. In addition, we conduct an adaptive dynamic weighting strategy to distribute the weight of different tasks automatically. Experiments on benchmark data sets demonstrate that the proposed MTDA-GNN outperforms traditional GNNs in graph-based semi-supervised node classification.

Abstract text summarization is a classic sequence-to-sequence natural language generation task. In order to improve the quality of unsupervised abstract text summarization in unsupervised mode, we propose two constraints for training text summarization model, embedding space constraint and information ratio constraint. We construct a generative adversarial network with two discriminators based on these two constraints (TC-SUM-GAN). We use unsupervised and supervised methods to train the model in the experiment. Experimental results show that the ROUGE-1 value of the unsupervised TC-SUM-GAN increases by $12.57$ points compared with the basic model and at least 1.96 points compared with other comparative models. The ROUGE scores of the supervised TC-SUM-GAN are also improved. TC-SUM-GAN achieves very competitive results for the metrics of ROUGE-1 and ROUGE-2. In addition, the abstracts generated by our model are closer to those generated manually.

This paper proposes LHFNet, an improved object detection method based on CenterNet, which achieves a better speed-accuracy trade-off. The main contributions of this paper are as follows: We design a lightweight Hourglass network to reduce the number of parameters and computations of CenterNet, thus improving the detection speed. We introduce an intermediate feature map fusion module to enhance the feature extraction capability of Hourglass, which compensates for the possible performance degradation caused by the network simplification. We apply an attention module to the final heat map to improve the image feature representation. We design and optimize a series of loss functions to train LHFNet effectively. Through the research in this paper, we demonstrate that the feature map fusion and the attention module greatly improve the performance of LHFNet. We evaluate our method on the COCO dataset and achieve a high accuracy at a fast speed (Tesla V100), which outperforms other object detection methods in terms of both accuracy and efficiency.

This paper presents a systematic literature review on optimizing feature extraction for palm and wrist multimodal biometrics. Identifying informative features across different modalities can be computationally expensive and time-consuming in such complex systems. Optimization techniques can streamline this process, making it more efficient thereby improving accuracy and reliability. The paper frames four research questions on input traits, approaches for feature extraction, classification approaches, and performance metrics of image data. The search query is generated based on the research questions that help retrieve the information on the above parameters. The focus of this paper is to provide the comprehensive and exhaustive gestalt of the appropriate input traits for image data from the information retrieved as well as optimal feature extraction and selection. However, the paper also intends to highlight the various classification approaches taken as well as the performance indicators against those classifiers. Further, the paper aims to analyze the effectiveness of various filtering techniques in eliminating image noise and improving overall system performance using MATLAB 2018. The paper concludes that a combination of palm and wrist biometrics could be a good input-trait combination. This work is novel as it covers multi-faceted processing, addressing various aspects of optimizing feature extraction and selection for palm and wrist multimodal biometrics.

Rolling bearing is a key component with the high fault rate in the rotary machines, and its fault diagnosis is important for the safe and healthy operation of the entire machine. In recent years, the deep learning has been widely used for the mechanical fault diagnosis. However, in the process of equipment operation, its state data always presents unbalanced. Number of effective data in different states is different and usually the gap is large, which makes it difficult to directly conduct deep learning. This paper proposes a new data enhancement method combining the resampling and Conditional Wasserstein Generative Adversarial Networks-Gradient Penalty (CWGAN-GP), and uses the gray images-based Convolutional Neural Network (CNN) to realize the intelligent fault diagnosis of rolling bearings. First, the resampling is used to expand the small number of samples to a large level. Second, the conditional label in Conditional Generative Adversarial Networks (CGAN) is combined with WGAN-GP to control the generated samples. Meanwhile, the Maximum Mean Discrepancy (MMD) is used to filter the samples to obtain the high-quality expanded data set. Finally, CNN is used to train the obtained dataset and carry out the fault classification. In the experiment, a single, compound and mixed fault cases of rolling bearings are successively simulated. For each case, the different sets considering the imbalance ratio of data are constructed, respectively. The results show that the method proposed significantly improves the fault diagnosis accuracy of rolling bearings, which provides a feasible way for the intelligent diagnosis of mechanical component with the complex fault modes and unbalanced small data.

Gene sequence classification can be seen as a challenging task due to the nonstationary, noisy and nonlinear characteristics of sequential data. The primary goal of this research is to develop a general solution approach for supervised DNA chromatogram (DNAC) classification in the absence of sufficient training data. Today, deep learning comes to the fore with its achievements, however this requires a lot of training data. Finding enough training data can be exceedingly challenging, particularly in the medical area and for rare disorders. In this paper, a novel supervised DNAC classification method is proposed, which combines three techniques to classify hepatitis virus DNA trace files as HBV and HCV. The features that are capable of reflecting the complex-structured sequential data are extracted based on both embedding and spectral entropies. After the supervised dimension reduction step, not only global behavior of the entropy features but also local behavior of the entropy features is taken into account for classification purpose. A memory-based learning, which cannot lose any information coming from training data as its nature, is being used as a classifier. Experimental results show that the proposed method achieves good results that although 19% training data is used, a performance of 92% is obtained.

In the field of medicine, thermal image processing and analysis play a significant role in the diagnosis, monitoring, and treatment of diseases. For example, during the last decade, several studies have been performed based on thermal image processing for ocular disease diagnosis. This research proposes a unique approach for the classification of subgroups of two retinal vascular diseases, namely diabetic eye disease and age-related macular degeneration (AMD). The class imbalance problem is a well-known issue when working with medical data, where one class is significantly less represented than another class in the dataset. To deal with the class imbalance issue, an ensemble decision tree classifier with a random under-sampling and adaptive boosting (RUSBoost) technique is proposed. The performance of the proposed classifier is compared with various traditional machine learning-based classifiers. Experimental results show that the proposed ensemble tree outperforms other classifiers through high accuracy, $F$-score, and Mathews correlation coefficient (MCC) values in classifying diabetic eye diseases and AMD diseases. The proposed ensemble decision tree distinguishes dry AMD and wet AMD over healthy controls with 95% average accuracy. Also, it classifies diabetic retinopathy (DR) with diabetic macular edema (DME) and DR without DME with 94% average accuracy. The classifier could distinguish dry and wet AMD which did not work around in temperature analysis on the manual temperature measurement. The performance of the automated classification model is on par with the performance of the temperature analysis of OST for DME and DR without DME.

This paper presents a new Data-Driven Teaching Learning-Based Optimization (DDTLBO) method used to improve the self-balancing and position control of a two-degree of freedom ball and balancer (2DOFBB) system. DDTLBO is a modified version of teaching learning-based optimization (TLBO) method, based on the natural pattern followed by “learners” in an academic class to improve their knowledge. In this, an intelligent algorithm is used on MATLAB platform to handle the nonlinearities and un-stability of the ball and balancer system. A ball-balancer is a highly nonlinear, electromechanical, multivariable and underactuated system. A sophisticated automated intelligent control approach is used to handle these complications. In the proposed approach, the pattern followed by inclination angle of balancer plate is traced after observing the control dynamics of 2DOFBB system. Modeling of ball balancer system is accomplished with the help of proportional integral and derivative (PID) controller. Further, an intelligent data-driven computational approach is implemented, which improves the execution of TLBO optimization method. This improved optimization is used to tune the performance of proportional integral derivative controller employed on the ball balancer system. This approach addresses the major challenge of unknown disturbances on the closed loop system of the considered problem. The dominance of proposed intelligent approach is verified over other optimization techniques using benchmark CEC functions. The optimized parameters of PID controllers hence obtained are implemented on 2DOFBB system. Results obtained are validated using real-time response analysis on the MATLAB/Simulation platform, which demonstrates the dominance of DDTLBO-PID controller over traditional controllers.