Query：
学者姓名：程航
Refining：
Year
Type
Indexed by
Source
Complex
Former Name
Co-
Language
Clean All
Abstract ：
由于细粒度图像之间存在小的类间方差和大的类内差异,现有分类算法仅仅聚焦于单张图像的显著局部特征的提取与表示学习,忽视了多张图像之间局部的异构语义判别信息,较难关注到区分不同类别的微小细节,导致学习到的特征缺乏足够区分度.本文提出了一种渐进式网络以弱监督的方式学习图像不同粒度层级的信息.首先,构建一个注意力累计目标定位模块(Attention accumulation object localization module, AAOLM),在单张图像上从不同的训练轮次和特征提取阶段对注意力信息进行语义目标集成定位.其次,设计一个多张图像异构局部交互图模块(Heterogeneous local interactive graph module, HLIGM),提取每张图像的显著性局部区域特征,在类别标签引导下构建多张图像的局部区域特征之间的图网络,聚合局部特征增强表示的判别力.最后,利用知识蒸馏将异构局部交互图模块产生的优化信息反馈给主干网络,从而能够直接提取具有较强区分度的特征,避免了在测试阶段建图的计算开销.通过在多个数据集上进行的实验,证明了提出方法的有效性,能够提高细粒度分类的精度.
Keyword ：
图神经网络 图神经网络 弱监督目标定位 弱监督目标定位 深度学习 深度学习 知识蒸馏 知识蒸馏 细粒度图像分类 细粒度图像分类
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 陈权 , 陈飞 , 王衍根 et al. 融合目标定位与异构局部交互学习的细粒度图像分类 [J]. | 自动化学报 , 2024 , 50 (11) : 2219-2230 . |
| MLA | 陈权 et al. "融合目标定位与异构局部交互学习的细粒度图像分类" . | 自动化学报 50 . 11 (2024) : 2219-2230 . |
| APA | 陈权 , 陈飞 , 王衍根 , 程航 , 王美清 . 融合目标定位与异构局部交互学习的细粒度图像分类 . | 自动化学报 , 2024 , 50 (11) , 2219-2230 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
高强度的高斯噪声往往会模糊或破坏图像的细节和结构,导致边缘信息的丢失.为此,提出基于逐像素强化学习的边缘保持图像复原算法.首先,为每个像素构建一个像素层智能体并设计针对边缘处的侧窗均值滤波器到动作空间中,所有的像素层智能体共享优势行动者-评论家算法的参数,因此模型可以同时输出所有位置的状态转移概率并选择合适的策略进行状态转移,从而复原图像;其次,在特征提取共享网络中结合协调注意力,聚焦所有像素位置在特征通道间的全局信息,并保留位置嵌入信息;然后,为了缓解稀疏奖励问题,设计一个基于图拉普拉斯正则的辅助损失,关注图像的局部平滑信息,对局部不平滑区域加以惩罚,从而促进像素层智能体更加有效地学习到正确的策略以实现边缘保持.实验结果表明,所提的算法在Middlebury2005数据集和MNIST数据集上的峰值信噪比(PSNR)分别达到32.97 dB和28.26 dB,相比于Pixel-RL算法分别提升了 0.23 dB和0.75 dB,参数量和训练总时间分别减少了 44.9％和18.2％,在实现边缘保持的同时有效降低了模型的复杂度.
Keyword ：
协调注意力 协调注意力 图像复原 图像复原 图拉普拉斯 图拉普拉斯 深度强化学习 深度强化学习 边缘保持 边缘保持 逐像素强化学习 逐像素强化学习
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 江敏 , 陈飞 , 程航 et al. 基于逐像素强化学习的边缘保持图像复原 [J]. | 计算机工程 , 2024 , 50 (12) : 224-232 . |
| MLA | 江敏 et al. "基于逐像素强化学习的边缘保持图像复原" . | 计算机工程 50 . 12 (2024) : 224-232 . |
| APA | 江敏 , 陈飞 , 程航 , 王美清 . 基于逐像素强化学习的边缘保持图像复原 . | 计算机工程 , 2024 , 50 (12) , 224-232 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Image Manipulation Localization (IML) is a fundamental binary segmentation task, focused on the precise identification and demarcation of the manipulated regions within an image. Most existing models primarily rely on RGB and noise features to accurately identify the tampered areas in images. However, in practical IML tasks, the effectiveness of commonly used noise feature modules is often compromised by the unknown tampering methods and the diversity of images. In this paper, we design an adaptive filter based on the Discrete Cosine Transform (DCT) to localize the manipulated regions. Furthermore, we theoretically demonstrate that the adaptive filter is equivalent to convolving the image with a large-scale convolutional kernel, thereby taking full account of features across the entire image. Through extensive experimentation on several benchmark image tampering datasets, our model has demonstrated performance that rivals the most state-of-the-art methods. © 2024 IEEE.
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wang, H. , Cheng, H. , Chen, J. et al. A Novel Adaptive DCT Filter for Image Manipulation Localization [未知]. |
| MLA | Wang, H. et al. "A Novel Adaptive DCT Filter for Image Manipulation Localization" [未知]. |
| APA | Wang, H. , Cheng, H. , Chen, J. , Xu, Y. . A Novel Adaptive DCT Filter for Image Manipulation Localization [未知]. |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The health-related Internet of Things (IoT) plays an irreplaceable role in the collection, analysis, and transmission of medical data. As a device of the health-related IoT, the electroencephalogram (EEG) has long been a powerful tool for physiological and clinical brain research, which contains a wealth of personal information. Due to its rich computational/storage resources, cloud computing is a promising solution to extract the sophisticated feature of massive EEG signals in the age of big data. However, it needs to solve both response latency and privacy leakage. To reduce latency between users and servers while ensuring data privacy, we propose a privacy-preserving feature extraction scheme, called LightPyFE, for EEG signals in the edge computing environment. In this scheme, we design an outsourced computing toolkit, which allows the users to achieve a series of secure integer and floating-point computing operations. During the implementation, LightPyFE can ensure that the users just perform the encryption and decryption operations, where all computing tasks are outsourced to edge servers for specific processing. Theoretical analysis and experimental results have demonstrated that our scheme can successfully achieve privacy-preserving feature extraction for EEG signals, and is practical yet effective.
Keyword ：
Additive secret sharing Additive secret sharing edge computing edge computing electroencephalogram (EEG) signal electroencephalogram (EEG) signal Internet of Things (IoT) Internet of Things (IoT) privacy-preserving privacy-preserving
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Yan, Nazhao , Cheng, Hang , Liu, Ximeng et al. Lightweight Privacy-Preserving Feature Extraction for EEG Signals Under Edge Computing [J]. | IEEE INTERNET OF THINGS JOURNAL , 2024 , 11 (2) : 2520-2533 . |
| MLA | Yan, Nazhao et al. "Lightweight Privacy-Preserving Feature Extraction for EEG Signals Under Edge Computing" . | IEEE INTERNET OF THINGS JOURNAL 11 . 2 (2024) : 2520-2533 . |
| APA | Yan, Nazhao , Cheng, Hang , Liu, Ximeng , Chen, Fei , Wang, Meiqing . Lightweight Privacy-Preserving Feature Extraction for EEG Signals Under Edge Computing . | IEEE INTERNET OF THINGS JOURNAL , 2024 , 11 (2) , 2520-2533 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Image denoising is a fundamental tool in the fields of image processing and computer vision. With the rapid development of multimedia and cloud computing, it has become popular for resource-constrained users to outsource the storage and denoising of massive images. However, it may cause privacy concerns and response delays. In this scenario, we propose an efFicient privAcy-preseRving Image deNoising schEme (FARINE) for outsourcing digital images. By introducing a key conversion mechanism, FARINE allows removing noise from a given noisy image using a non-local mean way without leaking any information about the plaintext content. Due to its low computational latency/communication cost, edge computing is considered to improve the user experience. To achieve a dynamic user set efficiently, we design a fine-grained access control mechanism to support user authorization and revocation in multi-user scenarios. Extensive experiments over several benchmark data sets show that FARINE obtains comparable performance to plaintext image denoising.
Keyword ：
access control access control edge computing edge computing homomorphic encryption homomorphic encryption image denoising image denoising Privacy-preserving Privacy-preserving
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Huang, Yibing , Xu, Yongliang , Cheng, Hang et al. Edge-based secure image denoising scheme supporting flexible user authorization [J]. | JOURNAL OF ALGORITHMS & COMPUTATIONAL TECHNOLOGY , 2024 , 18 . |
| MLA | Huang, Yibing et al. "Edge-based secure image denoising scheme supporting flexible user authorization" . | JOURNAL OF ALGORITHMS & COMPUTATIONAL TECHNOLOGY 18 (2024) . |
| APA | Huang, Yibing , Xu, Yongliang , Cheng, Hang , Chen, Fei , Wang, Meiqing . Edge-based secure image denoising scheme supporting flexible user authorization . | JOURNAL OF ALGORITHMS & COMPUTATIONAL TECHNOLOGY , 2024 , 18 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Existing vision-language pre-training models typically extract region features and conduct fine-grained local alignment based on masked image/text completion or object detection methods. However, these models often design independent subtasks for different modalities, which may not adequately leverage interactions between modalities, requiring large datasets to achieve optimal performance. To address these limitations, this paper introduces a novel pre-training approach that facilitates fine-grained vision-language interaction. We propose two new subtasks — image filling and text filling — that utilize data from one modality to complete missing parts in another, enhancing the model's ability to integrate multi-modal information. A selector mechanism is also developed to minimize semantic overlap between modalities, thereby improving the efficiency and effectiveness of the pre-trained model. Our comprehensive experimental results demonstrate that our approach not only fosters better semantic associations among different modalities but also achieves state-of-the-art performance on downstream vision-language tasks with significantly smaller datasets. © 2024 Elsevier Ltd
Keyword ：
Cross-modal Cross-modal Image captioning Image captioning Partial auxiliary Partial auxiliary Pre-training Pre-training
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Cheng, H. , Ye, H. , Zhou, X. et al. Vision-language pre-training via modal interaction [J]. | Pattern Recognition , 2024 , 156 . |
| MLA | Cheng, H. et al. "Vision-language pre-training via modal interaction" . | Pattern Recognition 156 (2024) . |
| APA | Cheng, H. , Ye, H. , Zhou, X. , Liu, X. , Chen, F. , Wang, M. . Vision-language pre-training via modal interaction . | Pattern Recognition , 2024 , 156 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Due to the existence of small inter-class differences and large intra-class variance among fine-grained images, the existing classification algorithms only focus on the extraction and representation learning of salient local features of a single image, ignoring the local heterogeneous semantic discrimination information between multiple images, difficult to pay attention to the subtle details that distinguish different categories, resulting in the lack of sufficient discrimination of the learned features. This paper proposes a progressive network to learn the information of different granularity levels of the image in a weakly supervised manner. First, attention accumulation object localization module (AAOLM) is constructed to perform semantic target integration localization on attention information from different training epochs and feature extraction stages on a single image. Second, a multi-image heterogeneous local interactive graph module (HLIGM) is designed to construct a graph network and aggregate information between the local region features of multiple images under the guidance of the category label after extracting the salient local region features of each image to enhance the discriminative power of the representation. Finally, the optimization information generated by HLIGM is fed back to the backbone by using knowledge distillation so that it can directly extract features with strong discrimination, avoiding the computational overhead of building the graph in the test phase. Through experiments on multiple data sets, it proves the effectiveness of the proposed method, which can improve the fine-grained classification accuracy. © 2024 Science Press. All rights reserved.
Keyword ：
Deep neural networks Deep neural networks Graph neural networks Graph neural networks Image enhancement Image enhancement Image representation Image representation Knowledge graph Knowledge graph Self-supervised learning Self-supervised learning Semantic Segmentation Semantic Segmentation Supervised learning Supervised learning
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Chen, Quan , Chen, Fei , Wang, Yan-Gen et al. Fine-grained Image Classification by Integrating Object Localization and Heterogeneous Local Interactive Learning [J]. | Acta Automatica Sinica , 2024 , 50 (11) : 2219-2230 . |
| MLA | Chen, Quan et al. "Fine-grained Image Classification by Integrating Object Localization and Heterogeneous Local Interactive Learning" . | Acta Automatica Sinica 50 . 11 (2024) : 2219-2230 . |
| APA | Chen, Quan , Chen, Fei , Wang, Yan-Gen , Cheng, Hang , Wang, Mei-Qing . Fine-grained Image Classification by Integrating Object Localization and Heterogeneous Local Interactive Learning . | Acta Automatica Sinica , 2024 , 50 (11) , 2219-2230 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Due to enormous computing and storage overhead for well-trained Deep Neural Network (DNN) models, protecting the intellectual property of model owners is a pressing need. As the commercialization of deep models is becoming increasingly popular, the pre-trained models delivered to users may suffer from being illegally copied, redistributed, or abused. In this paper, we propose DeepDIST, the first end-to-end secure DNNs distribution framework in a black-box scenario. Specifically, our framework adopts a dual-level fingerprint (FP) mechanism to provide reliable ownership verification, and proposes two equivalent transformations that can resist collusion attacks, plus a newly designed similarity loss term to improve the security of the transformations. Unlike the existing passive defense schemes that detect colluding participants, we introduce an active defense strategy, namely damaging the performance of the model after the malicious collusion. The extensive experimental results show that DeepDIST can maintain the accuracy of the host DNN after embedding fingerprint conducted for true traitor tracing, and is robust against several popular model modifications. Furthermore, the anti-collusion effect is evaluated on two typical classification tasks (10-class and 100-class), and the proposed DeepDIST can drop the prediction accuracy of the collusion model to 10% and 1% (random guess), respectively.
Keyword ：
anti-collusion anti-collusion Deep neural networks Deep neural networks digital fingerprinting digital fingerprinting digital watermarking digital watermarking
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Cheng, Hang , Li, Xibin , Wang, Huaxiong et al. DeepDIST: A Black-Box Anti-Collusion Framework for Secure Distribution of Deep Models [J]. | IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY , 2024 , 34 (1) : 97-109 . |
| MLA | Cheng, Hang et al. "DeepDIST: A Black-Box Anti-Collusion Framework for Secure Distribution of Deep Models" . | IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 34 . 1 (2024) : 97-109 . |
| APA | Cheng, Hang , Li, Xibin , Wang, Huaxiong , Zhang, Xinpeng , Liu, Ximeng , Wang, Meiqing et al. DeepDIST: A Black-Box Anti-Collusion Framework for Secure Distribution of Deep Models . | IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY , 2024 , 34 (1) , 97-109 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Image steganography attempts to imperceptibly hide the secret image within the cover image. Most of the existing deep learning -based steganography approaches have excelled in payload capacity, visual quality, and steganographic security. However, they are difficult to losslessly reconstruct secret images from stego images with relatively large payload capacity. Recently, although some studies have introduced invertible neural networks (INNs) to achieve largecapacity image steganography, these methods still cannot reconstruct the secret image losslessly due to the existence of lost information on the output side of the concealing network. We present an INN -based framework in this paper for lossless image steganography. Specifically, we regard image steganography as an image super -resolution task that converts low -resolution cover images to high -resolution stego images while hiding secret images. The feature dimension of the generated stego image matches the total dimension of the input secret and cover images, thereby eliminating the lost information. Besides, a bijective secret projection module is designed to transform various secret images into a latent variable that follows a simple distribution, improving the imperceptibility of the secret image. Comprehensive experiments indicate that the proposed framework achieves secure hiding and lossless extraction of the secret image.
Keyword ：
Covert communication Covert communication Information security Information security Invertible neural networks Invertible neural networks Lossless steganography Lossless steganography
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wang, Tingqiang , Cheng, Hang , Liu, Ximeng et al. Lossless image steganography: Regard steganography as super-resolution [J]. | INFORMATION PROCESSING & MANAGEMENT , 2024 , 61 (4) . |
| MLA | Wang, Tingqiang et al. "Lossless image steganography: Regard steganography as super-resolution" . | INFORMATION PROCESSING & MANAGEMENT 61 . 4 (2024) . |
| APA | Wang, Tingqiang , Cheng, Hang , Liu, Ximeng , Xu, Yongliang , Chen, Fei , Wang, Meiqing et al. Lossless image steganography: Regard steganography as super-resolution . | INFORMATION PROCESSING & MANAGEMENT , 2024 , 61 (4) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Although federated learning can provide privacy protection for individual raw data, some studies have shown that the shared parameters or gradients under federated learning may still reveal user privacy. Differential privacy is a promising solution to the above problem due to its small computational overhead. At present, differential privacy-based federated learning generally focuses on the trade-off between privacy and model convergence. Even though differential privacy obscures sensitive information by adding a controlled amount of noise to the confidential data, it opens a new door for model poisoning attacks: attackers can use noise to escape anomaly detection. In this paper, we propose a novel model poisoning attack called Model Shuffle Attack (MSA), which designs a unique way to shuffle and scale the model parameters. If we treat the model as a black box, it behaves like a benign model on test set. Unlike other model poisoning attacks, the malicious model after MSA has high accuracy on test set while reducing the global model convergence speed and even causing the model to diverge. Extensive experiments show that under FedAvg and robust aggregation rules, MSA is able to significantly degrade performance of the global model while guaranteeing stealthiness.
Keyword ：
Differential privacy Differential privacy Federated learning Federated learning Model poisoning Model poisoning Privacy-preserving Privacy-preserving
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Yang, Ming , Cheng, Hang , Chen, Fei et al. Model poisoning attack in differential privacy-based federated learning [J]. | INFORMATION SCIENCES , 2023 , 630 : 158-172 . |
| MLA | Yang, Ming et al. "Model poisoning attack in differential privacy-based federated learning" . | INFORMATION SCIENCES 630 (2023) : 158-172 . |
| APA | Yang, Ming , Cheng, Hang , Chen, Fei , Liu, Ximeng , Wang, Meiqing , Li, Xibin . Model poisoning attack in differential privacy-based federated learning . | INFORMATION SCIENCES , 2023 , 630 , 158-172 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
