Showing papers in "Pattern Recognition Letters in 2022"
••
TL;DR: In this article , a deep learning model for detection and prevention of attack in the cloud platform is presented, which is carried out in three phases like at first, Hidden Markov Model (HMM) is incorporated for the detection of attacks.
44 citations
••
TL;DR: In this paper , a novel unsupervised learning framework based on concept-based and hierarchical clustering is proposed for Twitter sentiment analysis, and two different feature representation methods including Boolean and Term frequency-inverse document frequency (TF-IDF) are investigated.
43 citations
••
TL;DR: In this paper , a hybrid active contour model driven by pre-fitting energy with an adaptive edge indicator function and an adaptive sign function is proposed to enable the evolution curve to adjust evolution direction and speed.
26 citations
••
TL;DR: In this article , a semi-supervised CycleGAN (SSA- CycleGAN) and Inception V3 transfer learning model has been developed to train the algorithm for detecting COVID-19 pandemic.
23 citations
••
TL;DR: In this article , IoU-balanced loss functions consisting of IoUbalanced classification loss and IoU balanced localization loss are proposed to solve the localization accuracy problems. But the proposed methods are not suitable for single-stage detectors.
22 citations
••
TL;DR: Wang et al. as discussed by the authors proposed a novel image encryption method based on logistic chaotic systems and deep autoencoder, which can effectively resist attacks and has excellent encryption performance while providing high security.
22 citations
••
TL;DR: Zhang et al. as discussed by the authors designed a hybrid backbone Res-Transformer based on ResNet-50 and Transformer block for effective identify information, which used global self-attention in place of depth-wise convolution in the fourth layer.
19 citations
••
TL;DR: This article proposed an unsupervised contrastive learning framework that is motivated from the perspective of label smoothing, which uses a novel contrastive loss that naturally exploits a data augmentation scheme in which new samples are generated by mixing two data samples with a mixing component.
19 citations
••
TL;DR: In this article , a sentiment analysis model is proposed to analyze the real time tweets, which are related to coronavirus, and the obtained feature vectors are fed to the ensemble classifier (GRU) and Capsule Neural Network (CapsNet)) for classifying the user's sentiment's as anger, sad, joy, and fear.
16 citations
••
TL;DR: Wang et al. as discussed by the authors proposed a novel Transformer-based Cross Reference Network (TCRN), which fully exploited long-range context dependencies in both feature representation extraction and cross-modal integration.
13 citations
••
TL;DR: In this paper , a generative adversarial network (GAN) based solution was proposed to generate the IR equivalent of a given visible image by training a deep network to learn the relation between visible and IR modalities.
••
TL;DR: Wang et al. as discussed by the authors proposed a video pre-processing technique called Temporal Sequence Sampling (TSS) for 2D face PAD by removing the estimated inter-frame 2D affine motion in the view and encoding the appearance and dynamics of the resulting smoothed video sequence into a single RGB image.
••
TL;DR: A comparative analysis of unimodal and multimodal behavioral biometric traits acquired while the subjects perform different activities on the phone, considering the touchscreen and the simultaneous background sensor data (accelerometer, gravity sensor, gyroscope, linear accelerometer, and magnetometer).
••
TL;DR: HuMIdb as discussed by the authors is a publicly available HCI database for mobile passive authentication, which includes multimodal behavioral biometric traits acquired while the subjects perform different activities on the phone such as typing, scrolling, drawing a number, and tapping on the screen.
••
TL;DR: Zhang et al. as discussed by the authors proposed a Super-Resolution guided Knowledge Distillation (SRKD) framework, which consists of two sub-networks: one is the super-resolution sub-network used to enhance the features of low-resolution images, and the other is the knowledge distillation sub network used to minimize the difference between the feature of high resolution images and the images output by the superresolution subnetwork.
••
TL;DR: The Compact Expression Recognition Network (CERN) as mentioned in this paper is a low-calorie network with only 1.45M parameters, which is almost 50x less than that of a state-of-the-art (SOTA) architecture.
••
TL;DR: Wang et al. as discussed by the authors proposed a context-related video anomaly detection method combined with a generative adversarial network, which first generated the video content with some continuous frames before and after this frame by using a two-branch generator network, and then minimized the prediction errors between the generated frames and its ground truth.
••
TL;DR: Wang et al. as mentioned in this paper proposed an improved S2ANET-SR model based on S2A-NET network, where the original and reduced image are fed to the detection network at the same time, and then a super-resolution enhancement module for the reduced image is designed to enhance the feature extraction of small objects, after that, the perceptual loss and texture matching loss is proposed as supervision.
••
TL;DR: In this article , a framework named IDSGT-DNN is proposed to improve the security in the cloud IDS system, which incorporates an attacker and a defender mechanism for attack and normal data processing.
••
TL;DR: In this paper , an indigenously developed X-band CW radar is employed to create a diverse DIAT-μ RadHAR dataset, which includes (a) army marching, (b) Stone pelting/Grenades throwing, (c) jumping with holding a gun, (d) army Jogging, (e) army crawling and (f) boxing activities.
••
TL;DR: Li et al. as mentioned in this paper proposed a two head contraction expansion convolutional neural network (CNN) architecture for robust presentation attack detection, which consists of raw image and edge enhanced image to learn discriminating features for binary classification.
••
TL;DR: Zhang et al. as mentioned in this paper investigated the merits of fusing multiple anomaly classifiers using weighted averaging (WA) fusion and proposed a novel three-stage optimization method with the following contributions: (a) a new hybrid optimisation method using Genetic Algorithm (GA) and Pattern Search (PS) to explore the weight space more effectively (b) a novel two-sided score normalisation method to improve the anomaly detection performance (c) a different ensemble pruning method to increase the generalisation performance.
••
TL;DR: Wang et al. as discussed by the authors proposed three weighting models based on tf-idf, k-nearest neighbor (kNN) based cosine similarity, and correlation score.
••
TL;DR: Wang et al. as discussed by the authors proposed a new data augmentation method for few-shot medical diagnosis, which utilizes both generative adversarial network and U-Net to generate diverse images.
••
TL;DR: In this paper , the authors proposed Deep Residual Spatiotemporal Translation Network (DR-STN), a novel unsupervised deep residual conditional GAN model with an Online Hard Negative Mining (OHNM) approach, which provides a wider network to learn a mapping from spatial to temporal representations and enhance the perceptual quality of synthesized images from a generator.
••
TL;DR: The authors proposed an end-to-end transformer-based approach to jointly perform text transcription and named entity recognition in handwritten documents, which achieved the state-of-the-art performance in the ICDAR 2017 Information Extraction competition.
••
TL;DR: In this paper , an evolutionary algorithm was developed to explore the hyper-parameters space of the IRCNN-VD and an improved regional convolution neural network was used to detect the vehicles.
••
TL;DR: In this article , the most recent advances accomplished in this field are reviewed, and a critical analysis on the current state of the art, as well as on the issues still open, is provided.
••
TL;DR: In this article , a novel automated model is proposed for an effective online product sentiment analysis, in which image and text normalization techniques are used to improve the data quality and feature extraction is performed by utilizing Latent Semantic Analysis (LSA), Term Frequency- Inverse Document Frequency (TF-IDF), Modified Local Binary Pattern (MLBP), and Speeded Up Robust Features (SURF) descriptors for extracting the textual and visual feature vectors from the preprocessed data.