检索结果-内蒙古大学图书馆

RIE-GAN: A retrievable image encryption method based on GAN

DIGITAL signal processing 2025年 163卷

作者： Nan, Yang Wo, Yan South China Univ Technol Sch Comp Sci & Engn Guangzhou Peoples R China

With the development of cloud computing, people tend to encrypt images and upload them to the cloud for saving storage space and protecting privacy. However, these image encryption methods will hinder the availability of images such as similarity retrieval. To address the issue, this paper proposes a retrievable image encryption method based on GAN(RIE-GAN), which ensures the high security and good retrieval performance of the ciphertext. RIE-GAN uses the convolutional neural network to extract image feature and utilizes the trained weight vector from the Group Normalization (GN) layer to evaluate the importance of each channel for retrieval and divide the feature into two subsets of different importance, the important subset for retrieval and the non-important subset. To achieve similarity retrieval of the encrypted important subset, we utilize Variable Feature Thumbnail Preserving Encryption (VF-TPE) to ensure that the mean value within a block of the encrypted important subset remains unchanged for retrieval. To further enhance security, we use a three-dimensional Lorenz chaotic system to encrypt the non-important subset. The two encrypted subsets are then merged into a whole, serving as the target domain for Cycle-GAN. By training the Cycle-GAN, we perform the feature transformation from the source domain to the target domain, thus ensuring both the security and retrieval functionality of the encrypted data. Experimental results on the Corel-1000 dataset demonstrate that our method can reach 0.886 in the accuracy of ciphertext retrieval, significantly outperforming other method and the security of the ciphertext is close to that of other encryption methods.

关键词： image similarity retrieval image encryption Cycle-GAN Feature division

来源：评论

学校读者我要写书评

暂无评论

Weakly supervised skin lesion segmentation based on spot-seeds guided optimal regions

引用

IET image processing 2023年第1期17卷 239-255页

作者： Al-Huda, Zaid Yao, Yuan Yao, Jing Peng, Bo Raza, Ali Southwest Jiaotong Univ Sch Comp & Artificial Intelligence Chengdu 610031 Sichuan Peoples R China Mfg Ind Chains Collaborat & Informat Support Tech Chengdu Peoples R China Sichuan Univ Dept Gen Practice West China Hosp Chengdu Sichuan Peoples R China Govt Coll Univ Faisalabad Dept Elect Engn & Technol Faisalabad Pakistan

Automatic skin lesion segmentation is the most critical and relevant task in computer-aided skin cancer diagnosis. methods based on convolutional neural networks (CNNs) are mainly used in current skin lesion segmentation. The requirement of huge pixel-level labels is a significant obstacle to achieve semantic segmentation of skin lesion by CNNs. In this paper, a novel weakly supervised framework for skin lesion segmentation is presented, which generates high-quality pixel-level annotations and optimizes the segmentation network. A hierarchical image segmentation algorithm can predict a boundary map for training images. Then, the optimal regions of candidate hierarchical levels are selected. Afterward, Superpixels-CRF built on the optimal regions is guided by spot seeds to propagate information from spot seeds to unlabeled regions, resulting in high-quality pixel-level annotations. Using these high-quality pixel-level annotations, a segmentation network can be trained and segmentation masks can be predicted. To iteratively optimize the segmentation network, the predicted segmentation masks are refined and the segmentation network are retrained. Comparative experiments demonstrate that the proposed segmentation framework reduces the gap between weakly and fully supervised skin lesion segmentation methods, and achieves state-of-the-art performance while reducing human labeling efforts.

关键词： current skin lesion segmentation computer-aided skin cancer diagnosis Biology and medical computing image segmentation spot seeds image classification skin learning (artificial intelligence) spot-seeds guided optimal regions predicted segmentation masks segmentation framework semantic segmentation automatic skin lesion segmentation neural nets Optical, image and video signal processing Computer vision and image processing techniques convolutional neural nets cancer candidate hierarchical levels convolutional neural networks weakly skin lesion segmentation methods high-quality pixel-level annotations huge pixel-level labels hierarchical image segmentation algorithm Patient diagnostic methods and instrumentation segmentation network medical image processing fully supervised skin lesion segmentation methods

来源：评论

学校读者我要写书评

暂无评论

FusionLungNet: Multi-scale fusion convolution with refinement network for lung CT image segmentation

引用

BIOMEDICAL signal processing AND CONTROL 2025年 107卷

作者： Rezvani, Sadjad Fateh, Mansoor Jalali, Yeganeh Fateh, Amirreza Shahrood Univ Technol Fac Comp Engn Shahrood Iran Iran Univ Sci & Technol IUST Sch Comp Engn Tehran Iran

Early detection of lung cancer is crucial as it increases the chances of successful treatment. Automatic lung image segmentation assists doctors in identifying diseases such as lung cancer, COVID-19, and respiratory disorders. However, lung segmentation is challenging due to overlapping features like vascular and bronchial structures, along with pixel-level fusion of brightness, color, and texture. New lung segmentation methods face difficulties in identifying long-range relationships between image components, reliance on convolution operations that may not capture all critical features, and the complex structures of the lungs. Furthermore, semantic gaps between feature maps can hinder the integration of relevant information, reducing model accuracy. Skip connections can also limit the decoder's access to complete information, resulting in partial information loss during encoding. To overcome these challenges, we propose a hybrid approach using the FusionLungNet network, which has a multi-level structure with key components, including the ResNet50 encoder, Channel-wise Aggregation Attention (CAA) module, Multi-scale Feature Fusion (MFF) block, self refinement (SR) module, and multiple decoders. The refinement sub-network uses convolutional neural networks for image post-processing to improve quality. Our method employs a combination of loss functions, including SSIM, IOU, and focal loss, to optimize image reconstruction quality. We created and publicly released a new dataset for lung segmentation called LungSegDB, including 1800 CT images from the LIDC-IDRI dataset (dataset version 1) and 700 images from the Chest CT Cancer images from Kaggle dataset (dataset version 2). Our method achieved an IOU score of 98.04, outperforming existing methods and demonstrating significant improvements in segmentation accuracy. Both the dataset and code are publicly available (Datasetlink, Codelink).

关键词： Lung CT image Segmentation Self refinement Hybrid loss function Feature fusion module

来源：评论

学校读者我要写书评

暂无评论

Utilizing Deep Convolutional neural Networks for Multi-Classification of Plant Diseases from image Data

引用

TRAITEMENT DU signal 2023年第4期40卷 1479-1490页

作者： Elumalai, Saraswathi Hussain, Faritha Banu Jahir SRM Inst Sci & Technol Dept Comp Sci & Engn Ramapuram CampusBharathi Salai Chennai 600089 Tamil Nadu India

The economic health of a nation is significantly influenced by the productivity of its agricultural sector. Enhancing this productivity is directly linked to the early detection and management of plant diseases. Automated classification methodologies are instrumental in the early diagnosis of these diseases, offering improved precision over traditional methods. These automated systems initiate disease detection as soon as symptoms begin to manifest on plant leaves, following a four-step process involving pre-processing, segmentation, feature extraction, and classification. In this study, we present an automated methodology for the detection and classification of plant diseases using a deep-learning approach applied to varying quality leaf images. A deep convolutional neural network architecture was trained utilizing an image dataset. The proposed Deep neural Network Plant Disease Classifier (DNN-PDC) was specifically designed for the multi-categorization of plant diseases. Tomato leaf images from the PlantVillage dataset on Kaggle were selected for the experiments. The proposed deep learning system demonstrated a high level of accuracy in the classification of various tomato leaf diseases, including Early Blight, Septoria Leaf Spot, and Late Blight. Experimental results indicate that the proposed method surpasses existing approaches in the image-based classification of tomato plant diseases. This study underscores the potential of the DNN-PDC model as a highly effective tool for plant disease detection and classification.

关键词： activation function convolutional neural network deep learning feature extraction image processing plant disease classification PlantVillage dataset segmentation

来源：评论

学校读者我要写书评

暂无评论

Single image defogging with a dual multiscale neural network model

引用

signal image AND VIDEO processing 2023年第4期17卷 1643-1651页

作者： Wang, Yong Li, Xinchao Jilin Univ Coll Commun Engn Changchun 130012 Peoples R China

images captured by image acquisition systems in scenes with fog or haze contain missing details, dull color, and reduced brightness. To address this problem, the dual multiscale neural network model based on the AOD theory is proposed in this paper. First, two parameters, namely transmittance and atmospheric layer coefficient, of the atmospheric scattering model are combined into a single parameter. The new neural network model proposed in this paper is then used to train this parameter. The network model proposed in this paper consists of two multiscale modules and a mapping module. In order to extract more perfect image features, this paper designs two multiscale modules for feature extraction. The convolution parameters of Multiscale Module 1 are designed to maintain the size of original images during feature extraction by adding pooling, sampling, etc. After each convolution operation, multiscale module 2 uses multiple small-sized convolution kernels for convolution, in which the concat operation is added to better connect the individual kernels, the mapping module maps the fogged images onto the extracted feature map and is able to extract more detail from the original image to obtain better defogging results after processing. Training is performed to derive a unified parameter model for image defogging, and finally, the defogged image is obtained using this parameter estimation model. The experimental results show that the model proposed this paper not only outperforms the AOD network in terms of peak signal-to-noise ratio, structural similarity, and subjective vision but also outperforms the mainstream deep learning and traditional methods in terms of image defogging;moreover, the defogged images are optimized in terms of detail, color, and brightness. In addition, ablation experiments had demonstrated that all of the structures in this paper were necessary.

关键词： neural network Deep learning Atmospheric scattering model

来源：评论

学校读者我要写书评

暂无评论

High efficiency deep image compression via channel-wise scale adaptive latent representation learning

引用

signal processing-image COMMUNICATION 2025年 130卷

作者： Wu, Chenhao Wu, Qingbo Ngan, King Ngi Li, Hongliang Meng, Fanman Xu, Linfeng Univ Elect Sci & Technol China Sch Informat & Commun Engn 2006 Xiyuan Ave West Hitech Zone Chengdu 611731 Sichuan Peoples R China

Recent learning based neural image compression methods have achieved impressive rate-distortion (RD) performance via the sophisticated context entropy model, which performs well in capturing the spatial correlations of latent features. However, due to the dependency on the adjacent or distant decoded features, existing methods require an inefficient serial processing structure, which significantly limits its practicability. Instead of pursuing computationally expensive entropy estimation, we propose to reduce the spatial redundancy via the channel-wise scale adaptive latent representation learning, whose entropy coding is spatially context- free and parallelizable. Specifically, the proposed encoder adaptively determines the scale of the latent features via a learnable binary mask, which is optimized with the RD cost. In this way, lower-scale latent representation will be allocated to the channels with higher spatial redundancy, which consumes fewer bits and vice versa. The downscaled latent features could be well recovered with a lightweight inter-channel upconversion module in the decoder. To compensate for the entropy estimation performance degradation, we further develop an inter-scale hyperprior entropy model, which supports the high efficiency parallel encoding/decoding within each scale of the latent features. Extensive experiments are conducted to illustrate the efficacy of the proposed method. Our method achieves bitrate savings of 18.23%, 19.36%, and 27.04% over HEVC Intra, along with decoding speeds that are 46 times, 48 times, and 51 times faster than the baseline method on the Kodak, Tecnick, and CLIC datasets, respectively.

关键词： Learned image compression Efficient decoding Scale-adaptive Inter-channel upconversion Inter-scale hyperprior

来源：评论

学校读者我要写书评

暂无评论

AUTOSGM: A UNIFIED LOWPASS REGULARIZATION FRAMEWORK FOR ACCELERATED LEARNING 49

AUTOSGM: A UNIFIED LOWPASS REGULARIZATION FRAMEWORK FOR ACCE...

引用

49th IEEE International Conference on Acoustics, Speech, and signal processing (ICASSP)

作者： Somefun, Oluwasegun A. Lee, Stefan Mathews, V. John Oregon State Univ Sch Elect Engn & Comp Sci Corvallis OR 97331 USA

ISBN: (纸本)9798350344868;9798350344851

This paper unifies commonly used accelerated stochastic gradient methods (Polyak's Heavy Ball, Nesterov's Accelerated Gradient and Adaptive Moment Estimation (Adam)) as specific cases of a general lowpass regularized learning framework, the Automatic stochastic Gradient Method (AutoSGM). For AutoSGM, we derive an optimal iteration-dependent learning rate function and realize an approximation. Adam is also an approximation of this optimal approach that replaces the iteration-dependent learning-rate with a constant. Empirical results on deep neural networks comparing the learning behavior of AutoSGM equipped with this iteration-dependent learning-rate algorithm demonstrate fast learning behavior, robustness to the initial choice of the learning rate, and can tune an initial constant learningrate in applications where a good constant learning rate approximation is unknown.

关键词： stochastic gradient method accelerated learning learning algorithms optimization deep learning.

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Deep Triplet Hashing for image Retrieval

引用

IEEE signal processing LETTERS 2024年 31卷 1489-1493页

作者： Meng, Lingtao Zhang, Qiuyu Yang, Rui Huang, Yibo Lanzhou Univ Technol Sch Comp & Commun Lanzhou 730050 Peoples R China Northwest Normal Univ Coll Phys & Elect Engn Lanzhou 730070 Peoples R China

Deep hashing enhances image retrieval accuracy by integrating hash encoding with deep neural networks. However, existing unsupervised deep hashing methods primarily rely on the rotational invariance of images to construct triplets, resulting in triplets that are unsatisfactory in both reliability and quantity. Additionally, some methods fail to adequately consider the relative similarity information between samples. To overcome these limitations, we propose a novel unsupervised deep triplet hashing method for image retrieval (abbreviated as UDTrHash). UDTrHash utilizes the extremal cosine similarity of deep features of images to construct more reliable first type triplets and expands the formed triplets through data augmentation strategies to introduce a larger number of triplets. Furthermore, we design a new triplet loss function to enhance the discriminative ability of the generated hash codes. Extensive experiments demonstrate that UDTrHash exhibits superior performance on three public benchmark datasets such as MIRFlickr25K compared to existing state-of-the-art hashing methods.

关键词： Data augmentation image retrieval triplets unsupervised deep hashing

来源：评论

学校读者我要写书评

暂无评论

Research on methods of Pavement Distress Detection using Convolutional neural Network based on Highway Rapid Inspection images

Research on Methods of Pavement Distress Detection using Con...

引用

9th International Conference on signal and image processing (ICSIP)

作者： Shu, Donglin Deng, Wenhao Li, Zibing Zhao, Chihang Wu, Jialun Zhao, Yong Zhang, Ziyi Huang, Yaxin Anhui Prov Expressway Testing & Testing Res Ctr C Yongnian Hebei Peoples R China Southeast Univ Sch Transportat Nanjing Peoples R China Anhui Prov Expressway Testing & Testing Res Ctr C Hefei Peoples R China

ISBN: (纸本)9798350350920

Aiming at the problem of intelligent detection of pavement distress based on highway rapid inspection images, this paper studies the intelligent detection technology of pavement distress based on Convolutional neural Networks (CNN). Firstly, the methods of pavement distress detection based on Faster R-CNN, SSD and RetinaNet are compared and analyzed. Secondly, three variants of CNN models are investigated for pavement distress detection of highway rapid inspection images, including Faster R-CNN-PDD-HRII, SSD-PDD-HRII and RetinaNet-PDD-HRII. Finally, the comparative experiments were conducted using SEU-BH dataset, and the results showed that the average of Faster R-CNN-RSDD-HRII is superior to the other two methods, with an average accuracy of 94.88% and F1-Score of 90.29%.

关键词： Convolutional neural Networks (CNN) Pavement distress Highway Rapid Inspection images Intelligent detection

来源：评论

学校读者我要写书评

暂无评论

Spectral Scaling-Based Augmentation for Corruption-Robust image Classification

引用

IEEE signal processing LETTERS 2025年 32卷 1206-1210页

作者： Zhang, Zhuang Zhang, Lijun Meng, Dejian Tian, Wei Yan, Jun Tongji Univ Sch Automot Studies Shanghai 201804 Peoples R China Tongji Univ Sch Elect & Informat Engn Shanghai 201804 Peoples R China

image classifiers often degrade in performance when test images differ significantly from the training distribution due to real-world image corruptions. Frequency-based augmentations can be used to address this issue, but existing methods excel against corruptions caused by noise and blur while struggling with those caused by contrast and fog. To tackle these challenges, we propose a novel image augmentation method grounded in a new perspective of relative spectral differences. This perspective characterizes spectral variations introduced by common corruptions as changes in non-zero frequencies, providing a unified understanding of their effects on image spectra. Building on this insight, the proposed method incorporates two key modules: a random spectral scaling module that captures statistical properties of image spectra and a deep spectral scaling module that adaptively learns spectral adjustments through a neural network. Experiments demonstrate that the proposed method improves overall robustness across various corruptions, with notable gains of 6.3% and 6.4% on contrast and fog, respectively, where existing methods often fall short.

关键词： Training Transforms Robustness image augmentation image color analysis Feature extraction Frequency-domain analysis Zirconium Pipelines image classification image corruption data augmentation image spectrum image classification

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：