检索结果-内蒙古大学图书馆

Enhancing out-of-distribution learning in computer vision through dominant feature masking

PATTERN ANALYSIS AND applications 2025年第2期28卷 1-30页

作者： Pilzak, Artem Thivierge, Jean-Philippe Univ Ottawa Sch Psychol Ottawa ON K1N 6N5 Canada Univ Ottawa Brain & Mind Res Inst Ottawa ON K1N 6N5 Canada

Out-of-distribution (OOD) learning presents a major challenge in machine learning as models must effectively generalize to previously unseen data. This challenge is prevalent in deep learning models, which tend to focus on the most dominant features in images. This narrow focus impedes OOD learning, where critical features are concealed or absent during testing, leading to reduced prediction accuracy. To address this issue, we introduce a novel data augmentation approach termed Dominant Feature Masking (DFM), inspired by human visual holistic processing. DFM strategically conceals and reveals the most prominent features within images, allowing neural networks to simultaneously capture both dominant and non-dominant attributes, thereby enhancing adaptability to OOD data. We evaluated DFM using a novel set of learning challenges termed Versatile Evaluation Benchmark (VEB), which assesses model performance on three distinct tasks: (i) augmented MNIST images to test resilience against diverse transformations;(ii) a novel dataset of unseen image classes to examine performance on new instances within familiar categories;and (iii) a dataset created by DALL-E to challenge class differentiation with artificially mixed features. Our results demonstrate that DFM significantly improves OOD generalization compared to traditional augmentation techniques, achieving marked enhancements across various conditions without compromising in-distribution testing accuracy. These findings underscore the potential of DFM to improve the performance of computer vision systems in various real-world scenarios, making them more robust and adaptable to unexpected data variations. By leveraging VEB, researchers will gain a deeper understanding of their models' generalization performance, ensuring that CNNs are well-equipped to handle the complexities of real-world applications. The source code and VEB datasets are available at https://***/Deepvisionary/DFM.

关键词： Out-of-distribution learning Convolutional neural networks Data augmentation Feature acquisition Domain generalization

来源：评论

学校读者我要写书评

暂无评论

Application of computer vision techniques to fermented foods: An overview

引用

TRENDS IN FOOD SCIENCE & TECHNOLOGY 2025年 160卷

作者： Song, Zheli Li, Yuanbo Zhao, Hongyuan Liu, Xiaogang Ding, Hailong Ding, Qiansu Ma, Dongna Liu, Shuangping Mao, Jian Jiangnan Univ Sch Artificial Intelligence & Comp Sci Wuxi 214122 Jiangsu Peoples R China Jiangnan Univ Natl Engn Res Ctr Cereal Fermentat & Food Biomfg Sch Food Sci & Technol State Key Lab Food Sci & Technol Wuxi 214122 Jiangsu Peoples R China Jiangnan Univ Shaoxing Key Lab Tradit Fermentat Food & Human Hlt Shaoxing Ind Technol Res Inst Shaoxing 312000 Zhejiang Peoples R China Luzhou Laojiao Grp Co Ltd Luzhou 646000 Sichuan Peoples R China Univ Minnesota Twin Cities Dept Agron & Plant Genet St Paul MN USA

Background: Fermented foods are products processed through microbial fermentation and are widely appreciated by consumers around the world for their unique flavors. With advancements in industrial technology and increasing consumer demand, modern techniques are being progressively integrated into the production and quality control of fermented foods to enhance production efficiency and product quality. Among these innovations, computer vision technology stands out as particularly impactful. Scope and approach: This paper provides an overview of the applications of computer vision in the field of fermented foods, focusing on its technical algorithms and applications within the food industry. It outlines the specific uses of computer vision technology across different types of fermented foods and discusses the relevant techniques employed. Finally, this review highlights the transformative potential of adaptive learning and multimodal fusion in addressing current limitations of computer vision for fermented food monitoring. Key findings and conclusions: The adoption of computer vision technology has significantly improved both the efficiency and accuracy of quality control processes in fermented food production. Through non-contact real-time monitoring, researchers can quickly identify the dynamic changes in microorganisms and related parameter indicators during fermentation and evaluate their impact on food quality. These technologies have not only boosted the efficiency of fermented food production but have also enhanced control over product flavor and safety assessments. Despite ongoing challenges in technology implementation and data analysis, the continuous advancements in deep learning and image processing technologies are expected to increase the impact of computer vision in the field of fermented foods, driving sustainable industry development.

关键词： Computer vision Fermented foods Food quality evaluation Nondestructive testing machine learning

来源：评论

学校读者我要写书评

暂无评论

Accelerating Brillouin fiber sensing via destructive-interference-enabled precise raw data acquisition and nonredundant image denoising

引用

OPTICA 2025年第2期12卷 216-227页

作者： Li, Zonglei Zhou, Yin Hu, Jianqi Yao, Jianping Yan, Lianshan Southwest Jiaotong Univ Ctr Informat Photon & Commun Sch Informat Sci & Technol Chengdu 610031 Sichuan Peoples R China Southwest Jiaotong Univ Sch Informat Sci & Technol Lab Intelligent Percept & Smart Operat & Maintenan Chengdu 610031 Sichuan Peoples R China Sorbonne Univ Coll France CNRS Lab Kastler BrosselENS Univ PSL 24 Rue Lhomond F-75005 Paris France Univ Ottawa Sch Elect Engn & Comp Sci Microwave Photon Res Lab 25 Templeton St Ottawa ON K1N 6N5 Canada

Distributed Brillouin fiber sensing, based on the linear relationship between Brillouin frequency shift (BFS) and physical quantities applied to sensing fibers, has found numerous applications in the past few decades. Recently, various advanced image denoising methods have been used for performance enhancements in Brillouin fiber sensors. Yet, even though these methods do significantly remove noises contained in raw data, the BFS measurement uncertainty is not reduced-the newly introduced image denoising appears redundant with the conventional signal processing. Here, in order to truly make Brillouin fiber sensing benefit from image denoising, we directly map BFS from the image-denoised data via the slope-assisted analysis of the Brillouin phase-gain ratio. As such, noise reduction resulting from image denoising fully translates into measurement uncertainty reduction. In order to further optimize the performance of image-denoising-enhanced Brillouin fiber sensing, we improve the quality of the raw Brillouin gain and phase data by designing an advanced coherent detection scheme called a microwave-photonic interferometer, which converts some amplitude and phase noises into common-mode noises and further eliminates them through destructive interference. A more than 20-fold sensing speed acceleration compared to the state-of-the-art is experimentally achieved. This remarkable performance enhancement is obtained by only optimizing the signal detection and processing unit, without modifying Brillouin scattering between pump and probe waves. Our method seamlessly connects Brillouin fiber sensing with advanced image denoising methods developed for computer vision and artificial intelligence, and makes imagedenoising-enhanced Brillouin fiber sensing outperform the state-of-the art significantly. (c) 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

关键词： Brillouin scattering Fiber optic sensors Heterodyne detection machine vision Phase noise Raman scattering

来源：评论

学校读者我要写书评

暂无评论

High precision banana variety identification using vision transformer based feature extraction and support vector machine

引用

SCIENTIFIC REPORTS 2025年第1期15卷 1-16页

作者： Ergun, Ebru Recep Tayyip Erdogan Univ Fac Engn & Architecture Dept Elect & Elect Engn Rize Turkiye

Bananas, renowned for their delightful flavor, exceptional nutritional value, and digestibility, are among the most widely consumed fruits globally. The advent of advanced image processing, computer vision, and deep learning (DL) techniques has revolutionized agricultural diagnostics, offering innovative and automated solutions for detecting and classifying fruit varieties. Despite significant progress in DL, the accurate classification of banana varieties remains challenging, particularly due to the difficulty in identifying subtle features at early developmental stages. To address these challenges, this study presents a novel hybrid framework that integrates the vision Transformer (ViT) model for global semantic feature representation with the robust classification capabilities of Support Vector machines. The proposed framework was rigorously evaluated on two datasets: the four-class BananaimageBD and the six-class BananaSet. To mitigate data imbalance issues, a robust evaluation strategy was employed, resulting in a remarkable classification accuracy rate (CAR) of 99.86%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\pm\:$$\end{document}0.099 for BananaSet and 99.70%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\pm\:$$\end{document}0.17 for BananaimageBD, surpassing traditional methods by a margin of 1.77%. The ViT model, leveraging self-supervised and semi-supervised learning mechanisms, demonstrated exceptional promise in extracting nuanced features critical for agricultural applications. By combining ViT features with cutting-edge machine learning classifiers, the proposed system establishes a ne

关键词： Agricultural diagnostics Banana classification Deep learning machine learning Precision agriculture vision transformer

来源：评论

学校读者我要写书评

暂无评论

Multichannel Object Detection with Event Camera 6

Multichannel Object Detection with Event Camera

引用

6th IEEE International Conference on image processing, applications and Systems, IPAS 2025

作者： Iliasov, Rafael Golkar, Alessandro Technical University of Munich Spacecraft Systems Munich Germany

ISBN: (纸本)9798331506520

object detection based on event vision has been a dynamically growing field in computer vision for the last 16 years. In this work, we create multiple channels from a single event camera and propose an event fusion method (EFM) to enhance object detection in event-based vision systems. Each channel uses a different accumulation buffer to collect events from the event camera. We implement YOLOv7 for object detection, followed by a fusion algorithm. Our multichannel approach outperforms single-channel-based object detection by 0.7% in mean Average Precision (mAP) for detection overlapping ground truth with IOU = 0.5. © 2025 IEEE.

关键词： machine vision

来源：评论

学校读者我要写书评

暂无评论

Advanced thermal vision techniques for enhanced fault diagnosis in electrical equipment: a review

引用

INTERNATIONAL JOURNAL OF SYSTEM ASSURANCE ENGINEERING AND MANAGEMENT 2025年第5期16卷 1914-1932页

作者： Sasithradevi, A. Persiya, J. Roomi, S. Mohamed Mansoor Perumal, D. Arumuga Prakash, P. Vijayalakshmi, M. Ebenezer, L. Brighty Vellore Inst Technol Ctr Adv Data Sci Chennai Tamil Nadu India Vellore Inst Technol Sch Elect Engn Chennai Tamil Nadu India Thiagarajar Coll Engn Dept Elect & Commun Engn Madurai Tamilnadu India Natl Inst Technol Karnataka Dept Mech Engn Surathkal Mangalore India Anna Univ Dept Elect Engn MIT Campus Chennai India

Ensuring the reliability and safety of electrical equipment is essential for industrial and residential applications. Traditional fault diagnosis methods involving physical inspections are time-consuming and ineffective for early fault detection. Infrared (IR) thermography offers a non-invasive and efficient solution by identifying anomalies in temperature profiles. This review explores thermal vision-based fault diagnosis techniques, including region of interest (ROI) segmentation, image pre-processing, and fault diagnosis algorithms, with a focus on deep learning approaches. The study highlights the effectiveness of machine learning models in enhancing fault detection accuracy while identifying challenges such as environmental variations, data inconsistencies, and system integration issues. The review discusses the role of real-time applications, wireless technologies, and AI-based automation in improving fault detection. Research gaps are identified, and future directions are proposed to enhance efficiency, reliability, and industrial adoption.

关键词： Fault diagnosis Infrared thermography Electrical equipment machine learning Deep learning Segmentation

来源：评论

学校读者我要写书评

暂无评论

Automatic visual recognition, detection and classification of weeds in cotton fields based on machine vision

引用

CROP PROTECTION 2025年 187卷

作者： Memon, Muhammad Sohail Chen, Shuren Shen, Baoguo Liang, Runzhi Tang, Zhong Wang, Shuai Zhou, Weiwei Memon, Noreena Jiangsu Univ Key Lab Modern Agr Equipment & Technol Minist Educ Zhenjiang 212013 Jiangsu Peoples R China Jiangsu Univ Sch Agr Engn Zhenjiang 212013 Jiangsu Peoples R China Sindh Agr Univ Fac Agr Engn Dept Farm Power & Machinery Tandojam 70060 Pakistan Jiangsu Aviat Tech Coll Zhenjiang Key Lab UAV Applicat Technol Zhenjiang 212134 Peoples R China

Crops and weeds are involved in a continuous competition for equal resources, which may result in a potential decrease in crop yields by up to 31% and an increase in the costs of agricultural inputs by up to 22% of cultivation. Weeds further impact crop production, and their detection is crucial for effective crop management. In this research, we targeted common weeds of cotton field, specifically i) Digitaria sanguinalis (L.) Scop, ii) Amaranthus retroflexus L., iii) Acalypha australis, L., iv) Cephalanoplos segetum, and v) Chenopodium album L. Additionally, image processing techniques such as grayscale conversion, binarization, and Gaussian and morphological filters were also utilized. These methods are based on machine vision and facilitate rapid and straightforward weed detection by segmenting, scrutinizing, and comparing input images. The plant height and area were obtained during cotton planting within 32 days and fitted to develop the growth law concerning planting days for achieving the function of distinguishing cotton from weeds. We conducted recognition experiments by dividing images into four quadrants and categorizing weeds as either inter-row or intra-row. Meanwhile, the inter-row planting information was used to identify weeds, and the leaf pixel area and circularity were used as the identification methods for intra-row weeds, which reduced the algorithm's running time and improved real-time performance. The experimental results indicated that the inter-row weed recognition rate was 89.4%, with an average processing time of 102ms. Whereas in the case of intra-row weeds, the recognition rate was measured at 84.6%, and the overall recognition rate for cotton was 85.0%, with a mean time consumption of 437ms. Furthermore, the present research underscores recent advancements such as machine vision and high-resolution imaging, which have significantly improved the accuracy of automated weed identification in cotton fields while acknowledging ongoing challen

关键词： Weed detection Inter-row weeds Intra-row weeds machine vision algorithms Weed segmentation Cotton crop Precision farming

来源：评论

学校读者我要写书评

暂无评论

3D Information in Robot vision System Based on Artificial Neural Network 14th

3D Information in Robot Vision System Based on Artificial Ne...

引用

14th International Conference on Frontier Computing, FC 2024

作者： Liu, Xiaoxiao Department of Mechanical and Electrical Engineering Jinan Engineering Vocational Technical College Shandong Jinan China

ISBN: (纸本)9789819627974

In the exploration of robot vision systems based on artificial neural networks, the research mainly focuses on their applications in 3D information recognition and processing. By simulating the processing of the human visual system, this technology enables robots to more effectively interpret and understand the three-dimensional spatial information of their environment, which has a revolutionary role in robot navigation, object recognition, obstacle avoidance and the execution of complex tasks. And this technology shows great potential in many fields such as industrial automation, autonomous vehicles, drone monitoring, and service robots. Not only that, it also plays a very important role in providing in-depth information, accurate positioning and efficient decision-making applications. According to the experimental data, among the 800 experimental subjects, more than 754 people were satisfied with the recognition accuracy, recognition speed, machine efficiency improvement, image recognition clarity, and overall satisfaction with the system. These findings indicate that with the continuous advancement of technology, 3D vision systems based on artificial neural networks will show more significant performance and value in future applications. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： machine vision

来源：评论

学校读者我要写书评

暂无评论

IMOFC: Identity-Level Metric Optimized Feature Compression for Identification Tasks

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2025年第2期35卷 1855-1869页

作者： Gao, Changsheng Jiang, Yiheng Wu, Siqi Ma, Yifan Li, Li Liu, Dong Univ Sci & Technol China Key Lab Brain Inspired Intelligent Percept & Cogni MOE Hefei 230037 Peoples R China Univ Missouri Dept Elect Engn Columbia MO 65211 USA

Feature compression has attracted much attention in recent years due to its promising applications in scenarios where features are transmitted and analyzed by machine vision. However, existing research mainly focuses on coarse-grained features extracted from recognition tasks such as classification and detection, neglecting fine-grained features extracted from identification tasks. In this paper, we make a pioneering attempt to study fine-grained feature compression in the context of identification tasks. Our main focus is on the distortion metric, given its critical importance in optimizing the performance of a compression network. We initiate our discussion by reviewing the instance-level metrics in existing literature, highlighting their oversight of the inter-feature relationships. The inter-feature relationships are especially important for identification tasks as they involve similarity comparison among different identities. To address this problem, we propose to consider inter-feature relationships from the perspective of identity information. Specifically, we propose an identity-level metric to incorporate both intra-identity similarity and inter-identity discriminability. The intra-identity similarity constraint aims to cluster features from the same identity, while the inter-identity discriminability constraint ensures that features from different identities deviate from each other. We implement the identity-level metric on four different feature compression networks designed based on feature characteristics. Experimental results show the effectiveness of the proposed identity-level metric on person re-identification and face verification tasks.

关键词： image coding Distortion Feature extraction Distortion measurement Semantics image reconstruction machine vision Feature compression identification tasks identity-level metric inter-feature relationships inter-identity discriminability

来源：评论

学校读者我要写书评

暂无评论

Toward Integrating ChatGPT Into Satellite image Annotation Workflows: A Comparison of Label Quality and Costs of Human and Automated Annotators

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2025年 18卷 4366-4381页

作者： Beck, Jacob Kemeter, Lukas Malte Duerrbeck, Konrad Abdalla, Mohamed Hesham Ibrahim Kreuter, Frauke Ludwig Maximilians Univ Munchen Munich Ctr Machine Learning MCML Inst Informat D-80538 Munich Germany Fraunhofer Inst Integrated Circuits IIS Ctr Appl Res Supply Chain Serv D-90411 Nurnberg Germany

High-quality annotations are a critical success factor for machine learning (ML) applications. To achieve this, we have traditionally relied on human annotators, navigating the challenges of limited budgets and the varying task-specific expertise, costs, and availability. Since the emergence of large language models (LLMs), their popularity for generating automated annotations has grown, extending possibilities and complexity of designing an efficient annotation strategy. Increasingly, computer vision capabilities have been integrated into general-purpose LLMs like ChatGPT. This raises the question of how effectively LLMs can be used in satellite image annotation tasks and how they compare to traditional annotator types. This study presents a comprehensive investigation and comparison of various human and automated annotators for image classification. We evaluate the feasibility and economic competitiveness of using the ChatGPT4-V model for a complex land usage annotation task and compare it with alternative human annotators. A set of satellite images is annotated by a domain expert and 15 additional human and automated annotators, differing in expertise and costs. Our analyzes examine the annotation quality loss between the expert and other annotators. This comparison is conducted through, first, descriptive analyzes, second, fitting linear probability models, and third, comparing F1-scores. Ultimately, we simulate annotation strategies where samples are split according to an automatically assigned certainty score. Routing low-certainty images to human annotators can cut total annotation costs by over 50% with minimal impact on label quality. We discuss implications regarding the economic competitiveness of annotation strategies, prompt engineering, and the task-specificity of expertise.

关键词： Annotations Satellite images Biological system modeling Costs Chatbots Training image annotation Visualization Remote sensing Data models Automated annotations ChatGPT label quality large language models (LLMs) satellite image annotation

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：