检索结果-内蒙古大学图书馆

Semantic Feature Integration network for Fine-grained Visual Classification

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Hui Li, Yueyang Luo, Haichi Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University 1800 Lihu Avenue Jiangsu Wuxi214000 China College of Internet of Things Engineering Jiangnan University 1800 Lihu Avenue Jiangsu Wuxi214000 China

Fine-Grained Visual Classification (FGVC) is known as a challenging task due to subtle differences among subordinate categories. Many current FGVC approaches focus on identifying and locating discriminative regions by using the attention mechanism, but neglect the presence of unnecessary features that hinder the understanding of object structure. These unnecessary features, including 1) ambiguous parts resulting from the visual similarity in object appearances and 2) noninformative parts (e.g., background noise), can have a significant adverse impact on classification results. In this paper, we propose the Semantic Feature Integration network (SFI-Net) to address the above difficulties. By eliminating unnecessary features and reconstructing the semantic relations among discriminative features, our SFI-Net has achieved satisfying performance. The network consists of two modules: 1) the multi-level feature filter (MFF) module is proposed to remove unnecessary features with different receptive field, and then concatenate the preserved features on pixel level for subsequent disposal;2) the semantic information reconstitution (SIR) module is presented to further establish semantic relations among discriminative features obtained from the MFF module. These two modules are carefully designed to be light-weighted and can be trained end-to-end in a weakly-supervised way. Extensive experiments on four challenging fine-grained benchmarks demonstrate that our proposed SFI-Net achieves the state-of-the-arts performance. Especially, the classification accuracy of our model on CUB-200-2011 and Stanford Dogs reaches 92.64% and 93.03%, respectively. Copyright © 2023, The Authors. All rights reserved.

关键词： Image classification

Infrared and visible image fusion using a deep learning framework

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Li, Hui Wu, Xiao-Jun Kittler, Josef Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University Wuxi214122 China CVSSP University of Surrey GuildfordGU2 7XH United Kingdom

In recent years, deep learning has become a very active research tool which is used in many image processing fields. In this paper, we propose an effective image fusion method using a deep learning framework to generate a single image which contains all the features from infrared and visible images. First, the source images are decomposed into base parts and detail content. Then the base parts are fused by weighted-averaging. For the detail content, we use a deep learning network to extract multi-layer features. Using these features, we use l1-norm and weighted-average strategy to generate several candidates of the fused detail content. Once we get these candidates, the max selection strategy is used to get the final fused detail content. Finally, the fused image will be reconstructed by combining the fused base part and the detail content. The experimental results demonstrate that our proposed method achieves state-of-the-art performance in both objective assessment and visual quality. The Code of our fusion method is available at https://***/hli1221/imagefusion_deeplearning. Copyright © 2018, The Authors. All rights reserved.

关键词： Image fusion

Locality constraint dictionary learning with support vector for pattern classification

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Yin, He-Feng Wu, Xiao-Jun Chen, Su-Gen School of Internet of Things Engineering Jiangnan University Wuxi214122 China Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University Wuxi214122 School of Mathematics and Computational Science Anqing Normal University Anqing246133

Discriminative dictionary learning (DDL) has recently gained significant attention due to its impressive performance in various pattern classification tasks. However, the locality of atoms is not fully explored in conventional DDL approaches which hampers their classification performance. In this paper, we propose a locality constraint dictionary learning with support vector discriminative term (LCDL-SV), in which the locality information is preserved by employing the graph Laplacian matrix of the learned dictionary. To jointly learn a classifier during the training phase, a support vector discriminative term is incorporated into the proposed objective function. Moreover, in the classification stage, the identity of test data is jointly determined by the regularized residual and the learned multi-class support vector machine. Finally, the resulting optimization problem is solved by utilizing the alternative strategy. Experimental results on benchmark databases demonstrate the superiority of our proposed method over previous dictionary learning approaches on both hand-crafted and deep features. The source code of our proposed LCDL-SV is accessible at https://***/yinhefeng/LCDL-SV. Copyright © 2019, The Authors. All rights reserved.

关键词： Support vector machines

Infrared and visible image fusion with ResNet and zero-phase component analysis

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Li, Hui Wu, Xiao-Jun Durrani, Tariq S. Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University Wuxi214122 China Department of Electronic and Electrical Engineering University of Strathclyde GlasgowG1 1XW United Kingdom

Feature extraction and processing are key tasks in the Image fusion algo- rithm, while most of deep learning-based methods use deep features directly without feature processing. This leads to the fusion performance degradation in some cases. To solve this drawback, in this paper, a novel fusion frame- work based on deep features and zero-phase component analysis (ZCA) is pro- posed. Firstly, the residual network (ResNet) is used to extract deep features from source images. Then ZCA and l1-norm are utilized to normalize the deep features and obtain initial weight maps. The final weight maps are obtained by employing a soft-max operation in association with the initial weight maps. Finally, the fused image is reconstructed using a weighted-averaging strategy. Compared with the existing fusion methods, experimental results demonstrate that the proposed algorithm achieves better performance in both objective assessment and visual quality. The code of our fusion algorithm is available at https://***/hli1221/imagefusion-resnet50. Copyright © 2018, The Authors. All rights reserved.

关键词： Infrared imaging

MoNET: No-reference image quality assessment based on a multi-depth output network

学校读者我要写书评

暂无评论

Journal of Electronic Imaging 2021年第4期30卷 043007-043007页

作者： Sang, Qingbing Su, Chenfei Zhu, Lingying Liu, Lixiong Wu, Xiaojun Bovik, Alan C. Jiangnan University School of Artificial Intelligence and Computer Science Wuxi China Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Wuxi China Beijing Institute of Technology School of Computer Science and Technology Beijing China University of Texas at Austin Laboratory for Image and Video Engineering AustinTX United States

When deep convolutional neural networks perform feature extraction, the features computed at each layer express different abstractions of visual information. The earlier layers extract highly compact low-level features such as bandpass and directional primitives, whereas deeper layers extract structural features of increasing abstraction, similar to contours, shapes, and edges, becoming less effable as the depth increases. We propose a different kind of end-to-end no-reference (NR) image quality assessment (IQA) model, which is defined as a multi-depth output convolutional neural network (MoNET). It accomplishes this by mapping both shallow and deep features to perceived quality. MoNET delivers three outputs that express shallow (lower-level) and deep (high-level) features, and maps them to subjective quality scores. The multiple outputs are combined into a single, final quality score. MoNET does this by combining the responses of three learning machines, so it may be viewed as a form of ensemble learning. The experimental results on three public image quality databases show that our proposed model achieves better performance than other state-of-the-art NR IQA algorithms. © 2021 SPIE and IS and T.

关键词： Convolutional neural networks

Non-negative representation based discriminative dictionary learning for face recognition

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Chen, Zhe Wu, Xiao-Jun Kittler, Josef Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence School of IoT Engineering Jiangnan University Wuxi214122 China CVSSP University of Surrey GuildfordGU2 7XH United Kingdom

In this paper, we propose a non-negative representation based discriminative dictionary learning algorithm (NRDL) for multicategory face classification. In contrast to traditional dictionary learning methods, NRDL investigates the use of non-negative representation (NR), which contributes to learning discriminative dictionary atoms. In order to make the learned dictionary more suitable for classification, NRDL seamlessly incorporates non-negative representation constraint, discriminative dictionary learning and linear classifier training into a unified model. Specifically, NRDL introduces a positive constraint on representation matrix to find distinct atoms from heterogeneous training samples, which results in sparse and discriminative representation. Moreover, a discriminative dictionary encouraging function is proposed to enhance the uniqueness of class-specific sub-dictionaries. Meanwhile, an inter-class incoherence constraint and a compact graph based regularization term are constructed to respectively improve the discriminability of learned classifier. Experimental results on several benchmark face data sets verify the advantages of our NRDL algorithm over the state-of-the-art dictionary learning methods. Copyright © 2019, The Authors. All rights reserved.

关键词： Face recognition

Class-specific residual constraint non-negative representation for pattern classification

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Yin, He-Feng Wu, Xiao-Jun Jiangnan University School of Internet of Things Engineering No. 1800 Lihu Avenue Wuxi214122 China Jiangsu Provincial Laboratory of Pattern Recognition and Computational Intelligence No. 1800 Lihu Avenue Wuxi214122 China

Representation based classification method (RBCM) remains one of the hottest topics in the community of pattern recognition, and the recently proposed non-negative representation based classification (NRC) achieved impressive recognition results in various classification tasks. However, NRC ignores the relationship between the coding and classification stages. Moreover, there is no regularization term other than the reconstruction error term in the formulation of NRC, which may result in unstable solution leading to misclassification. To overcome these drawbacks of NRC, in this paper, we propose a class-specific residual constraint non-negative representation (CRNR) for pattern classification. CRNR introduces a class-specific residual constraint into the formulation of NRC, which encourages training samples from different classes to competitively represent the test sample. Based on the proposed CRNR, we develop a CRNR based classifier (CRNRC) for pattern classification. Experimental results on several benchmark datasets demonstrate the superiority of CRNRC over conventional RBCM as well as the recently proposed NRC. Moreover, CRNRC works better or comparable to some state-of-the-art deep approaches on diverse challenging pattern classification tasks. The source code of our proposed CRNRC is accessible at https://***/yinhefeng/CRNRC. Copyright © 2019, The Authors. All rights reserved.

关键词： pattern recognition

Enhanced Long-Term Prediction Based on 2d Tensorization for Building Energy Consumption

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Kong, Jun Deng, Song Jiang, Min Zhuang, Danfeng Jiangnan University Wuxi214122 China Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University Wuxi214122 China IBISC Univ. Evry Université Paris-Saclay 91020 France

Building energy consumption constitutes a significant contributor to greenhouse gas emissions. Accurate prediction of building energy consumption is crucial for building energy management, thus contributing to the reduction of green-house gas emissions. However, existing prediction algorithms often focus on within-period variations in energy con-sumption data and overlook between-period variations. This limitation makes it challenging to achieve precise predic-tion results, especially for long-term forecasts. To address this research gap, this paper proposes an Enhanced Long-Term Prediction based on 2D Tensorization (ELP2T) for building energy consumption. First, a Frequency-Guided 2D Tensorization Network (FG-2TN) is proposed. Energy consumption data, when represented as a one-dimensional time series, faces certain limitations, and FG-2TN is employed to address these limitations. Second, a Pro-gress-Optimized Deep Convolutional Network (PO-DCN) is proposed. It is designed to efficiently extract and learn features from the obtained two-dimensional tensors with fewer parameters and less time.. Third, A modular method is proposed to transform one-dimensional time series into two-dimensional tensors using Fast Fourier transform. These tensors are processed and then concatenated into a one-dimensional sequence for output. Final, a comparative analy-sis was conducted with various conventional prediction algorithms to demonstrate the superiority of the ELP2T model. The obtained average R² score for our proposed method is 0.807, representing an impressive enhancement of 11.06% compared to the most advanced alternatives. This substantial improvement firmly establishes the superiority of our approach. Especially, when considering a prediction length of 720 units, the performance gain of this metric increases to 18.09%, underscoring the pronounced advantage of our method in addressing long-term forecasting scenarios. © 2023, The Authors. All rights reserved.

关键词： Fast Fourier transforms

Face recognition via compact second order image gradient orientations

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yin, He-Feng Wu, Xiao-Jun Song, Xiao-Ning Jiangnan University School of Artificial Intelligence and Computer Science No. 1800 Lihu Avenue Wuxi214122 China Jiangsu Provincial Laboratory of Pattern Recognition and Computational Intelligence No. 1800 Lihu Avenue Wuxi214122 China

Conventional subspace learning approaches based on image gradient orientations only employ the first-order gradient information. However, recent researches on human vision system (HVS) uncover that the neural image is a landscape or a surface whose geometric properties can be captured through the second order gradient information. The second order image gradient orientations (SOIGO) can mitigate the adverse effect of noises in face images. To reduce the redundancy of SOIGO, we propose compact SOIGO (CSOIGO) by applying linear complex principal component analysis (PCA) in SOIGO. Combined with collaborative representation based classification (CRC) algorithm, the classification performance of CSOIGO is further enhanced. CSOIGO is evaluated under real-world disguise, synthesized occlusion and mixed variations. Experimental results indicate that the proposed method is superior to its competing approaches with few training samples, and even outperforms some prevailing deep neural network based approaches. The source code of CSOIGO is available at https://***/yinhefeng/SOIGO. © 2022, CC BY-SA.

关键词： Face recognition