检索结果-内蒙古大学图书馆

Privacy-Preserving Autoencoder for Collaborative Object Detection

IEEE TRANSACTIONS ON image processing 2024年 33卷 4937-4951页

作者： Azizian, Bardia Bajic, Ivan V. Simon Fraser Univ Sch Engn Sci Burnaby BC V5A 1S6 Canada

Privacy is a crucial concern in collaborative machine vision where a part of a Deep Neural Network (DNN) model runs on the edge, and the rest is executed on the cloud. In such applications, the machine vision model does not need the exact visual content to perform its task. Taking advantage of this potential, private information could be removed from the data insofar as it does not significantly impair the accuracy of the machine vision system. In this paper, we present an autoencoder-style network integrated within an object detection pipeline, which generates a latent representation of the input image that preserves task-relevant information while removing private information. Our approach employs an adversarial training strategy that not only removes private information from the bottleneck of the autoencoder but also promotes improved compression efficiency for feature channels coded by conventional codecs like VVC-Intra. We assess the proposed system using a realistic evaluation framework for privacy, directly measuring face and license plate recognition accuracy. Experimental results show that our proposed method is able to reduce the bitrate significantly at the same object detection accuracy compared to coding the input images directly, while keeping the face and license plate recognition accuracy on the images recovered from the bottleneck features low, implying strong privacy protection. Our code is available at https://***/bardia-az/ppa-code.

关键词： image coding Data privacy Training Privacy Codecs Visualization machine vision Deep neural network coding for machines privacy model inversion attack collaborative intelligence adversarial training feature compression

来源：评论

学校读者我要写书评

暂无评论

A fast specular removal method for a single real image☆

引用

DISPLAYS 2025年 87卷

作者： Hao, Chuanpeng He, Yan Li, Yufeng Niu, Xiaobo Wang, Yan Chongqing Univ State Key Lab Mech Transmiss Adv Equipment Chongqing 400030 Peoples R China Univ Brighton Sch Comp Engn & Math Brighton BN2 4GJ England

The specular reflection of objects is an important factor affecting image display quality, which poses challenges to tasks such as pattern recognition and machine vision detection. At present, specular removal for a single real image is a crucial pre-processing step to improve the performance of computer vision algorithms. Despite notable approaches tailored for handling synthesized and pre-simplified images with dark backgrounds, real-time separation of specular reflection for a single real image remains a challenging problem. This paper proposes a novel specular removal method to separate the specular reflection for a single real image accurately and efficiently based on the dark channel prior. Initially, a modified-specular-free (MSF) image is developed using the dark channel prior, which can derive a direct estimation of specular reflection. Next, the image chromaticity spaces are established to represent the pixel intensity. Then, the maximum chromaticity value of the modified MSF image is extracted to guide the filtering of the specular reflection, treating the specular pixels as noise in the chromaticity space. Finally, the image without specular reflection can be obtained using the restored maximum chromaticity value based on the dichromatic reflection model. The superiority of this method is to achieve highquality specular reflection separation quickly without destroying the geometric features of the real image. Compared with the state-of-the-art methods, experimental results show that the proposed algorithm can achieve the best subjective visual effect and satisfactory quantitative performance. In addition, this approach can be implemented efficiently to meet real-time requirements, promising to be applied to computer vision measurement and inspection applications.

关键词： Specular removal Highlight Dark channel machine vision image restoration

来源：评论

学校读者我要写书评

暂无评论

Adaptive adam-based optimizers using second-order weight decoupling and gradient-aware weight decay for vision transformer

引用

machine vision AND applications 2025年第3期36卷 1-14页

作者： Sai, Boyapati Hemanth Mukherjee, Snehasis Dubey, Shiv Ram Shiv Nadar Inst Eminence Greater Noida UP India Indian Inst Informat Technol Allahabad Prayagraj UP India

Optimizers play important roles in enhancing the performance of a deep network. A study on different optimizers is necessary to understand the effect of optimizers on the performance of the deep network for a given target task, such as image classification. Several attempts were made to investigate the effect of optimizers on the performance of CNNs. However, such experiments have not been carried out on vision transformers (ViT), despite the recent success of ViT in various image processing tasks. In this paper, we conduct exhaustive experiments with ViT using different optimizers. In our experiments, we found that weight decoupling and weight decay in optimizers play important roles in training ViT. We focused on the concept of weight decoupling and tried different variations of it to investigate to what extent weight decoupling is beneficial for a ViT. We propose two techniques that provide better results than weight-decoupled optimizers: (i) The weight decoupling step in optimizers involves a linear update of the parameter with weight decay as the scaling factor. We propose a quadratic update of the parameter which involves using a linear as well as squared parameter update using the weight decay as the scaling factor. (ii) We propose using different weight decay values for different parameters depending on the gradient value of the loss function with respect to that parameter. A smaller weight decay is used for parameters with a higher gradient value and vice versa. image classification experiments are conducted over CIFAR-100 and TinyimageNet datasets to observe the performance of these proposed methods with respect to state-of-the-art optimizers such as Adam, RAdam, and AdaBelief. The code is available at https://***/Hemanth-Boyapati/Adaptive-weight-decay-optimizers.

关键词： Adam-based optimizer Weight decoupling Transformers Weight decay Adaptive optimizers

来源：评论

学校读者我要写书评

暂无评论

A hybrid transformer-sequencer approach for age and gender classification from in-wild facial images

引用

NEURAL COMPUTING & applications 2024年第3期36卷 1149-1165页

作者： Singh, Aakash Singh, Vivek Kumar Banaras Hindu Univ Dept Comp Sci Varanasi 221005 India

The advancements in computer vision and image processing techniques have led to emergence of new application in the domain of visual surveillance, targeted advertisement, content-based searching, human-computer interaction, etc. Out of the various techniques in computer vision, face analysis, in particular, has gained much attention. Several previous studies have tried to explore different applications of facial feature processing for a variety of tasks, including age and gender classification. However, despite several previous studies having explored the problem, the age and gender classification of in-wild human faces is still far from achieving the desired levels of accuracy required for real-world applications. This paper, therefore, attempts to bridge this gap by proposing a hybrid model that combines self-attention and BiLSTM approaches for age and gender classification problems. The proposed model's performance is compared with several state-of-the-art models proposed so far. An improvement of approximately 10% and 6% over the state-of-the-art implementations for age and gender classification, respectively, is noted for the proposed model. The proposed model is thus found to achieve superior performance and is found to provide a more generalized learning. The model can, therefore, be applied as a core classification component in various image processing and computer vision problems.

关键词： Age classification Gender classification Hybrid classification approach Sequencer vision transformer

来源：评论

学校读者我要写书评

暂无评论

Enhanced Classification System for Real-Time Embedded vision applications

引用

IEEE ACCESS 2024年 12卷 162311-162326页

作者： Khelifi, Ramzi Nini, Brahim Berkane, Mohamed Univ Oum El Bouaghi Res Lab Comp Sci Complex Syst ReLa CS 2 Oum El Bouaghi 04000 Algeria Univ Oum El Bouaghi Artificial Intelligence & Autonomous Things Lab Oum El Bouaghi 04000 Algeria

Embedded computer vision systems are increasingly being adopted across various domains, playing a pivotal role in enabling advanced technologies such as autonomous vehicles and industrial automation. Their cost-effectiveness, compact size, and portability make them particularly well-suited for diverse implementations and operations. In real-time scenarios, these systems must process visual data with minimal latency, which is crucial for immediate decision-making. However, these solutions continue to face significant challenges related to computational efficiency, memory usage, and accuracy. This research addresses these challenges by enhancing classification methodologies, specifically in Gray Level Co-occurrence Matrix (GLCM) feature extraction and Support Vector machine (SVM) classifiers. To maintain a high level of accuracy while preserving performance, a smaller feature set is selected following a comprehensive complexity analysis and is further refined through Correlation-based Feature Selection (CFS). The proposed method achieves an overall classification accuracy of 84.76% with a feature set reduced by 79.2%, resulting in a 72.45% decrease in processing time, a 50% reduction in storage requirements, and up to a 77.8% decrease in memory demand during prediction. These improvements demonstrate the effectiveness of the proposed approach in improving the adaptability and capabilities of embedded vision systems (EVS), optimizing their performance under the constraints of real-time limited-resource environments.

关键词： Accuracy Support vector machines Real-time systems Feature extraction Memory management Computer vision Surveillance Bandwidth Wildlife machine learning image processing Embedded computer vision limited resource systems machine learning pattern classification real-time image processing

来源：评论

学校读者我要写书评

暂无评论

Computer vision on X-Ray Data in Industrial Production and Security applications: A Comprehensive Survey

引用

IEEE ACCESS 2023年 11卷 2445-2477页

作者： Rafiei, Mehdi Raitoharju, Jenni Iosifidis, Alexandros Aarhus Univ DIGIT Dept Elect & Comp Engn Aarhus Denmark Univ Jyvaskyla Fac Informat Technol Jyvaskyla 40100 Finland

X-ray imaging technology has been used for decades in clinical tasks to reveal the internal condition of different organs, and in recent years, it has become more common in other areas such as industry, security, and geography. The recent development of computer vision and machine learning techniques has also made it easier to automatically process X-ray images and several machine learning-based object (anomaly) detection, classification, and segmentation methods have been recently employed in X-ray image analysis. Due to the high potential of deep learning in related image processing applications, it has been used in most of the studies. This survey reviews the recent research on using computer vision and machine learning for X-ray analysis in industrial production and security applications and covers the applications, techniques, evaluation metrics, datasets, and performance comparison of those techniques on publicly available datasets. We also highlight some drawbacks in the published research and give recommendations for future research in computer vision-based X-ray analysis.

关键词： X-ray imaging Security Computer vision Imaging Industrial engineering Three-dimensional displays Deep learning deep learning X-ray industrial applications security applications

来源：评论

学校读者我要写书评

暂无评论

vision-based monitoring of railway superstructure: A review

引用

CONSTRUCTION AND BUILDING MATERIALS 2024年 442卷

作者： Aela, Peyman Cai, Jiafu Jing, Guoqing Chi, Hung-Lin Hong Kong Polytech Univ Dept Bldg & Real Estate Hung Hom Hong Kong Peoples R China Beijing Jiaotong Univ Sch Civil Engn Beijing 100044 Peoples R China

The computer vision-based analysis of railway superstructure has gained significant attention in railway engineering. This approach utilises advanced image processing and machine learning techniques to extract valuable information from visual data captured in the railway track environment. By analysing images from various sources such as cameras, drones, or sensors, computer vision algorithms can accurately detect and classify different components of the ballast superstructure, including the catenary system support, rail surface and profile, fastening system, sleeper, and ballast layer. This enables the automated assessment of the railway track's condition, stability, and maintenance needs. This paper comprehensively reviews the recent advancements, challenges, and potential applications of computer vision techniques in analysing railway superstructure. It discusses various vision-based methodologies and machine-learning approaches utilised in this context. Furthermore, it examines the benefits and limitations of computer vision-based analysis and presents future research directions for improving its applicability in railway track engineering.

关键词： Railway superstructure Track inspection Computer vision machine learning Robotics

来源：评论

学校读者我要写书评

暂无评论

machine vision system for automatic defect detection of ultrasound probes

引用

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY 2024年第7-8期135卷 3421-3435页

作者： Profili, Andrea Magherini, Roberto Servi, Michaela Spezia, Fabrizio Gemmiti, Daniele Volpe, Yary Univ Florence Dept Ind Engn Via Santa Marta 3 I-50139 Florence Italy Esaote SpA Via Caciolle 15 I-50127 Florence Italy

Industry 4.0 conceptualizes the automation of processes through the introduction of technologies such as artificial intelligence and advanced robotics, resulting in a significant production improvement. Detecting defects in the production process, predicting mechanical malfunctions in the assembly line, and identifying defects of the final product are just a few examples of applications of these technologies. In this context, this work focuses on the detection of ultrasound probes' surface defects, with a focus on Esaote S.p.A.'s production line probes. To date, this control is performed manually and therefore biased by many factors such as surface morphology, color, size of the defect, and by lighting conditions (which can cause reflections preventing detection). To overcome these shortfalls, this work proposes a fully automatic machine vision system for surface acquisition of ultrasound probes coupled with an automated defect detection system that leverage artificial intelligence. The paper addresses two crucial steps: (i) the development of the acquisition system (i.e., selection of the acquisition device, analysis of the illumination system, and design of the camera handling system);(ii) the analysis of neural network models for defect detection and classification by comparing three possible solutions (i.e., MMSD-Net, ResNet, EfficientNet). The results suggest that the developed system has the potential to be used as a defect detection tool in the production line (full image acquisition cycle takes similar to 200 s), with the best detection accuracy obtained with the EfficientNet model being 98.63% and a classification accuracy of 81.90%.

关键词： Product characterization Inspection system image processing Artificial intelligence Virtual prototyping

来源：评论

学校读者我要写书评

暂无评论

A Regret Bound for the AdaMax Algorithm With image Segmentation Application

引用

MATHEMATICAL METHODS IN THE APPLIED SCIENCES 2025年第9期48卷 10208-10214页

作者： Jirakipuwapat, Wachirapong King Mongkuts Univ Technol North Bangkok KMUTNB Fac Sci Energy & Environm Rayong Thailand

The AdaMax algorithm provides enhanced convergence properties for stochastic optimization problems. In this paper, we present a regret bound for the AdaMax algorithm, offering a tighter and more refined analysis compared to existing bounds. This theoretical advancement provides deeper insights into the optimization landscape of machine learning algorithms. Specifically, the You Only Look Once (YOLO) framework has become well-known as an extremely effective object segmentation tool, mostly because of its extraordinary accuracy in real-time processing, which makes it a preferred option for many computer vision applications. Finally, we used this algorithm for image segmentation.

关键词： AdaMax deep learning image segmentation regret bound YOLO

来源：评论

学校读者我要写书评

暂无评论

Improve accuracy in CNNs while using approximate computing methods

引用

JOURNAL OF SUPERCOMPUTING 2025年第2期81卷 1-23页

作者： Rafieinejad, Mohammadreza Marvasti, Mohammadreza Binesh Asghari, Seyyed Amir Shahbakhti, Kimiya Kharazmi Univ Dept Elect & Comp Engn Tehran Iran

Artificial neural networks have been one of the science's most influential and essential branches in the past decades. Neural networks have found applications in various fields including medical and pharmaceutical services, voice and speech recognition, computer vision, natural language processing, and video and image processing. Neural networks have many layers and consume much energy. Approximate computing is a promising way to reduce energy consumption in applications that can tolerate a degree of accuracy reduction. This paper proposes an effective method to prevent accuracy reduction after using approximate computing methods in the CNNs. The method exploits the k-means clustering algorithm to label pixels in the first convolutional layer. Then, using one of the existing pruning methods, different pruning amounts have been applied to all layers. The experimental results on three CNNs and four different datasets show that the accuracy of the proposed method has significantly improved (by 17%) compared to the baseline network.

关键词： Approximate computing Computer vision Convolutional neural network Energy consumption machine learning Neural network

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：