检索结果-内蒙古大学图书馆

CNN-EFF: CNN Based Edge Feature Fusion in Semantic image Labelling and Parsing

NEURAL processing LETTERS 2022年第3期54卷 1753-1781页

作者： Srivastava, Vishal Biswas, Bhaskar IIT BHU Dept Comp Sci & Engn Varanasi Uttar Pradesh India Bennett Univ Dept Comp Sci & Engn Greater Noida India

Semantic segmentation and image parsing have rapidly become an eminent research area in computer vision and machine learning domain. Many applications have required a robust mechanism for segmentation, such as self-driving, augmentative reality, object recognition, etc. Due to the high applicability in the various domains, In this paper, we have introduced a two-step frame-work that parses the image into predefined labels by using a novel CNN architecture and improving the likelihood of labels. In step-1, nine-layer CNN architecture has been introduced, which trains on minimal training samples and results in the pixel-wise Soft-Max probabilities. These probabilities are the soft estimates derived from a hard classifier, i.e., MLP. Data in step-1 has been prepared in the form of a patch-label set. In step-2, we have introduced a Jacobian optimization-based label relaxation method that fuses the local extrema as an edge prior. The proposed frame-work has been denoted as CNN-EFF in this work. The CNN-EFF scheme has been evaluated two publicly available benchmark data-sets, which has arranged in the form of image and their pixel label ground-truth. The experimental results have been compared with the previously proposed state-of-the-art methods. The CNN-EFF has greatly improved semantic labeling accuracy up to a significant gain from the past techniques. The CNN-EFF process has reported 84.42%, 85.91%, 94.66%, 97.14%, and 98.27% accuracy for the Highway, House, sheep, Horse-rider, and Horse-keeper images, respectively. Conclusively, the Proposed frame-work has out-performed the previously proposed state-of-the-art methods.

关键词： Convolutional neural networks (CNNs) Deep-learning Jacobian optimization Semantic relaxation

来源：评论

学校读者我要写书评

暂无评论

Multi-person Pose Estimation with Multi-Attention Mechanism

Multi-person Pose Estimation with Multi-Attention Mechanism

引用

image processing, Computer vision and machine Learning (ICICML), International Conference on

作者： Qian Shao Yeqin Shao School of Transportation and Civil Engineering Nantong University Nantong China

ISBN: (数字)9798350355413

ISBN: (纸本)9798350355420

In recent years, multi-person pose estimation has emerged as a prominent research direction in the field of computer vision, holding significant importance in applications such as human-computer interaction, action analysis, and virtual reality. However, traditional methods are often complex and inefficient, particularly during the feature fusion process, which can lead to the loss of critical information and increased errors under occlusion and complex poses. To address this, multiple attention modules were introduced in the early stages of the network to enhance the modeling of dependencies between key points, ultimately overcoming the limitations of conventional heatmaps through a coordinate classification approach. The design utilizing multiple attention modules achieves a balance between maintaining a lightweight structure and improving accuracy. Furthermore, by introducing a multi-attention mechanism, information loss is reduced, thereby enhancing the model's robustness in handling occlusion and other challenges in complex scenarios. Compared to existing advanced methods, the approach presented in this paper achieves an average precision increase of 1.5 percentage points on the COCO dataset.

关键词： Heating systems Solid modeling Computer vision Accuracy Computational modeling Pose estimation Virtual reality Feature extraction Robustness Load modeling

来源：评论

学校读者我要写书评

暂无评论

Research on Target Defect Detection Based on machine vision 4

Research on Target Defect Detection Based on Machine Vision

引用

2021 4th International Conference on Modeling, Simulation and Optimization Technologies and applications, MSOTA 2021

作者： Lu, Chuqiao Li, Zhiyong Shi, Shaoyue Peng, Shaolong School of International Education Wuhan University of Technology Wuhan China School of Automation Wuhan University of Technology Wuhan China School of Navigation Wuhan University of Technology Wuhan China

The daily detection of highspeed electric multiple units (EMU) body is very important for China railway maintenance system. This paper proposes a new method based on machine vision to detect bolts and switches on EMU side skirt, which aims to replace manpower. Yolov3 network is used to identify and locate bolts and switches. After positioning, the status of bolts is detected through Alexnet network. Experiments show that the processing method can achieve the detection work efficiently and accurately. It is suitable for many cases such as insufficient illumination and non-vertical shooting of image samples, which has high robustness. © Published under licence by IOP Publishing Ltd.

关键词： Bolts

来源：评论

学校读者我要写书评

暂无评论

A novel offline handwritten text recognition technique to convert ruled-line text into digital text through deep neural networks

引用

MULTIMEDIA TOOLS AND applications 2022年第13期81卷 18223-18249页

作者： Qureshi, Faiza Rajput, Asif Mujtaba, Ghulam Fatima, Noureen Sukkur IBA Univ Ctr Excellence Robot Artificial Intelligence & Bl Dept Comp Sci Sukkur Pakistan

Offline Handwritten Text Recognition (HTR) has been an active area of research due to its wide range of applications and challenges. Recently, many offline HTR techniques have been developed. However, most of the existing techniques were trained on the datasets that contain the handwritten text images on plain pages. Nevertheless, in real life, the handwritten text can be written on either plain pages or ruled-line pages. Therefore, the approaches proposed in recent literature are unable to convert the digital text accurately written on ruled-line pages. Hence, this study proposes a tailor-made end-to-end offline HTR technique that can accurately convert the offline handwritten text written on ruled-line pages into digital text with the help of computer vision and deep neural network-based techniques. To Evaluate the performance of our proposed technique, we developed a relatively complex dataset that contains the hand-written text images on the ruled-line pages. Our experimental results show that our proposed technique is capable of converting the hand-written text on ruled-line pages into digital text with an overall accuracy of 76.7%. Moreover, the experimental results show that our proposed technique obtained 20% more accurate results compared to baseline techniques. We believe that our proposed technique will contribute positively in the body of knowledge in the field of offline HTR. Moreover, the modular design of our proposed technique allows tailored modifications with respect to data while eliminating the need to retrain the neural network-based models.

关键词： Offline hand-written text recognition Ruled-line handwritten text recognition Ruled-lines Deep learning machine learning Digital image processing

来源：评论

学校读者我要写书评

暂无评论

Computer vision-Based Hand Recognition and Gesture Control for Dino Games

Computer Vision-Based Hand Recognition and Gesture Control f...

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Balumuri Dinesh C V Naveeth Reddy N. Senthamilarasi Sathyabama Institute of Science and Technology Chennai India

ISBN: (数字)9798331515683

ISBN: (纸本)9798331515690

Hand Recognition and Gesture Control For Dino Game Using Computer vision to control the popular Chrome Dino game using hand recognition and gesture control through computer vision techniques. The system leverages real-time image processing and machine learning algorithms to detect and interpret hand gestures, allowing for an intuitive and interactive gaming experience. The proposed method utilizes a webcam to capture live video feed, from which hand landmarks are extracted using a pre-trained neural network model. Various hand gestures, such as swipe and hold, are then mapped to corresponding in-game actions such as jumping and ducking. This gesture-based control mechanism not only enhances user engagement but also demonstrates the potential of computer vision in creating touchless interfaces for gaming applications.

关键词： Hands Computer vision machine learning algorithms Sensitivity Webcams Scalability Lighting Games Streaming media Real-time systems

来源：评论

学校读者我要写书评

暂无评论

SPA: Self-Peripheral-Attention for central-peripheral interactions in endoscopic image classification and segmentation

引用

EXPERT SYSTEMS WITH applications 2024年 245卷

作者： Huo, Xiangzuo Tian, Shengwei Yang, Yongxu Yu, Long Zhang, Wendong Li, Aolun Xinjiang Univ Sch Comp Sci & Technol Urumqi 830000 Xinjiang Peoples R China Xinjiang Univ Xinjiang Key Lab Signal Detect & Proc Urumqi 830000 Xinjiang Peoples R China Xinjiang Med Univ Xinjiang Canc Ctr Key Lab Oncol Affiliated Tumour Hosp Urumqi 830011 Xinjiang Peoples R China

Peripheral vision is a vital component of human visual processing that allows for efficient and accurate recognition of visual features across diverse regions of the visual field. Analogously, endoscopic images often exhibit peripheral regions of blur, due to their inherent imaging properties. Previous strategies employing either coarse-grained global attention or fine-grained local attention to enhance performance have often inadvertently compromised the intrinsic self-attention mechanism of multilayer transformers, leading to less optimal solutions. This research introduces Self-Peripheral-Attention (SPA), an innovative mechanism that incorporates peripheral vision modeling into self-attention, so as to enhance the accuracy and efficiency of classification and segmentation tasks in endoscopic imaging. SPA synthesizes fine-grained central and coarsegrained peripheral interactions and possesses three primary characteristics: (i) peripheral contextualization aggregation;(ii) interaction between coarse-grained peripheral and fine-grained central features facilitated by depthwise dilated convolution;(iii) element-wise affine transformation to integrate attention into the value. The effectiveness and generalizability of the proposed SPA -Net were assessed on XJUEE, XJUEESEG, Kvasir and Kvasir-SEG endoscopy datasets. The results underscore the potential of peripheral vision modeling in self-attention for augmenting machine perception models. The associated code can be accessed at https://***/huoxiangzuo/SPA.

关键词： Peripheral vision Endoscopic image classification Endoscopic image segmentation Self-attention Feature fusion

来源：评论

学校读者我要写书评

暂无评论

Deep Learning and machine Learning Based Efficient Framework for image Based Plant Disease Classification and Detection

Deep Learning and Machine Learning Based Efficient Framework...

引用

2022 International Conference on Advanced Computing Technologies and applications, ICACTA 2022

作者： Nancy, P. Pallathadka, Harikumar Naved, Mohd Kaliyaperumal, Karthikeyan Arumugam, K. Garchar, Vipul Saveetha Institute of Medical and Technical Sciences Chennai India Manipur International University Manipur India Amity University Noida India AMBO University IT HH Campus India Karpagam Academy of Higher Education Coimbatore India Junagadh Agricultural University Junagadh India

ISBN: (纸本)9781665495158

Without agriculture, human existence would be inconceivable. A large percentage of the world's population relies on agriculture for their daily needs. In addition, it creates a big number of jobs in the area. Using traditional agricultural practices results in lower yields, which is the fault of farmers. Agriculture and allied sectors will continue to be critical to the economy's long-term growth and prosperity. Farming has a slew of challenges, including disease detection and control and crop monitoring and tracking. Farming with intelligence is a realistic option in many situations. Smart agriculture is now possible because to the internet of things and machine learning approaches. Computer vision, image processing, and machine learning techniques are used in the automated leaf disease diagnostic system to analyze photographs of diseased leaves. A farmer can make an educated choice regarding a plant illness thanks to automated disease detection equipment that speeds up the diagnostic process. A farmer had to first send the contaminated leaf to a pathology lab for confirmation of the illness, which was a tedious process. It is the purpose of this paper to propose a framework for the real-time classification of agricultural images. Crop disease pictures categorization and illness prediction are made easier using this system. © 2022 IEEE.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Segmentation of Smoke Plumes Using Fast Local Laplacian Filtering 1

引用

7th International Conference on Computer vision and image processing, CVIP 2022

作者： Koranne, Vedant Anand Ientilucci, Emmett J. Dey, Abhishek Datta, Aloke Ghosh, Susmita Rochester Institute of Technology Electrical Engineering RochesterNY United States Rochester Institute of Technology Center for Imaging Science RochesterNY United States Bethune College University of Calcutta Kolkata India The LNM Institute of Information Technology Jaipur India Jadavpur University Kolkata India

ISBN: (数字)9783031314179

ISBN: (纸本)9783031314162

In this paper, we address the problem of smoke plume segmentation from background clutter. Smoke plumes can be generated from fires, explosions, etc. In the mining industry, plumes from blasts need to be characterized in terms of their volume and concentration, for example. Plume segmentation is required in order to start such an analysis. We present a new image processing approach based on a fast local Laplacian filtering (FLLF) technique. In addition, we discuss how we designed and executed our own field experiments to acquire actual test data of smoke plumes from RGB video cameras. Lastly, we show how the FLLF technique can be used to generate thousands of training samples with applications in machine learning. Results show that the FLLF technique outperforms state-of-the-art approaches (i.e., SFFCM and an approach by Wang et al.) when tested using metrics such as Accuracy, the Jaccard Index, F1-score, False Alarms and Misses. We also show that the FLLF technique is more computationally efficient. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Robust Optimization-based Neural Architectures Against Adversarial Attacks

Robust Optimization-based Neural Architectures Against Adver...

引用

IEEE SSD International Multi-Conference on Systems, Signals and Devices

作者： Khaoula Ben Ali Seifeddine Messaoud Mohamed Ali Hajjaji Mohamed Atri Noureddine Liouane Laboratory of Automatic Signal Image Processing National Engineering School Monastir University Monastir Tunisia Faculty of Sciences of Monastir Monastir University Monastir Tunisia Higher Institute of Applied Sciences and Technology of Sousse Sousse University Sousse Tunisia Computer Engineering Department College of Computer Science King Khalid University Abha Saudi Arabia

ISBN: (数字)9798331542726

ISBN: (纸本)9798331542733

Recent years have seen a rapid development in machine Learning, which has profoundly influenced many areas of science and engineering. Among them, computer vision takes the leading place, where important tasks are image classifications powered by CNNs. Despite the great performance of CNNs in complicated scenarios, they remain sensitive to so-called adversarial attacks, and deliberate perturbations leading them to incorrect predictions. Besides more innocuous consequences, this has serious security implications for critical applications, in-cluding medical diagnostics, where misclassifications might result in disastrous outcomes. This research work discusses adversarial attacks on CNNs and other DNNs in computer vision, studying a full range of the generation and detection methods with details while discussing intrinsic vulnerability and robustness. It also proposes a learning framework that will enhance the robustness and security of DNNs and CNNs against such adversarial perils. The ultimate goal is directed to an improvement in the reliability of such models in absolutely critical scenarios for safe deployment into applications where accuracy is crucial.

关键词： Computer vision Accuracy Perturbation methods Computer network reliability machine learning Robustness Security Medical diagnosis Resilience image classification

来源：评论

学校读者我要写书评

暂无评论

Efficient Real-Time Okra Stage Identification using YOLOV8 10

Efficient Real-Time Okra Stage Identification using YOLOV8

引用

10th IEEE International Conference on Electronics, Computing and Communication Technologies (IEEE CONECCT)

作者： Kumar, Nikhilesh B. Ferbin, F. J. Sivapatham, Shoba Kar, Asutosh Krithiga, R. Vellore Inst Technol Sch Mech Engn Chennai Tamil Nadu India Vellore Inst Technol Ctr Adv Data Sci Chennai Tamil Nadu India Natl Inst Technol Elect & Commun Engn Jalandhar Punjab India Vellore Inst Technol Sch Comp Sci Chennai Tamil Nadu India

ISBN: (纸本)9798350385939;9798350385922

In modern agriculture, crop growth monitoring is a crucial component, as it offers intuitive information about the health and growth of the plant, assisting farmers and other agricultural specialists. This systematic growth monitoring is necessary for crop health and agricultural productivity. We preferred the YOLOv8, which utilizes machine learning and offers efficient plant analysis in agriculture. This preferred method predicts bounding boxes and the probability of each possible class, allowing it to achieve exceptional detection speed without trading off accuracy. Pre-processing was done on the created, "Okra-dataset" to standardize the image to a fixed resolution and enhance our dataset's strength. We tested our work using different models: YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x. Our test revealed that YOLOv8x achieved the highest mean average precision (mAP) of 82.9%. The implementation of research indicates that YOLOv8x is a good tool for agricultural applications, which can be very helpful to farmers.

关键词： Crop growth monitoring computer vision YOLOv8

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：