检索结果-内蒙古大学图书馆

Hybrid two-stage cascade for instance segmentation of overlapping objects

PATTERN ANALYSIS AND applications 2023年第3期26卷 957-967页

作者： Yang, Yakun Luo, Wenjie Tian, Xuedong Hebei Univ Sch Cyber Secur & Comp Baoding 071002 Peoples R China Hebei Univ Hebei Machine Vis Engn Res Ctr Baoding 071002 Peoples R China Hebei Univ Lab Intelligence Image & Text Baoding 071002 Peoples R China

Although two-stage methods of instance segmentation achieve better performance than one-stage counterparts, the segmentation results on overlapping objects are unsatisfactory. We found that occlusion significantly impacts the location of adjacent objects and produces coarse masks without adequate refinements. To circumvent the issue, we propose a hybrid model for instance segmentation called HTCIS, which iteratively forms the detection and segmentation. The main idea is to improve overall performance by optimizing every component based on a two-stage cascade structure. Compared with existing models, our approach decreases the loss of feature information, including semantic and detailed features. The detection branch prioritizes location accuracy when ranking bounding boxes, while the segmentation branch explores more contextual information and segments pixels in a multi-view fashion with the guide of an attention mechanism. Experimental results demonstrate that HTCIS is capable of processing occlusion. We conclude that multi-refinement of two-stage cascade is essential for accurate segmentation of overlapping objects, and our optimization is efficient in achieving this goal.

关键词： Computer vision Instance segmentation Two-stage cascade model Hybrid tasks learning Object occlusion

来源：评论

学校读者我要写书评

暂无评论

ComAI: Enabling Lightweight, Collaborative Intelligence by Retrofitting vision DNNs 41

ComAI: Enabling Lightweight, Collaborative Intelligence by R...

引用

41st IEEE Conference on Computer Communications (IEEE INFOCOM)

作者： Jayarajah, Kasthuri Wanniarachchige, Dhanuja Abdelzaher, Tarek Misra, Archan Univ Maryland Baltimore Cty Baltimore MD 21228 USA Singapore Management Univ Singapore Singapore Univ Illinois Champaign IL USA

ISBN: (纸本)9781665458221

While Deep Neural Network (DNN) models have transformed machine vision capabilities, their extremely high computational complexity and model sizes present a formidable deployment roadblock for AIoT applications. We show that the complexity-vs-accuracy-vs-communication tradeoffs for such DNN models can be significantly addressed via a novel, lightweight form of "collaborative machine intelligence" that requires only runtime changes to the inference process. In our proposed approach, called ComAI, the DNN pipelines of different vision sensors share intermediate processing state with one another, effectively providing hints about objects located within their mutually-overlapping Field-of-Views (FoVs). CoMAI uses two novel techniques: (a) a secondary shallow ML model that uses features from early layers of a peer DNN to predict object confidence values in the image, and (b) a pipelined sharing of such confidence values, by collaborators, that is then used to bias a reference DNN's outputs. We demonstrate that CoMAI (a) can boost accuracy (recall) of DNN inference by 20-50%, (b) works across heterogeneous DNN models and deployments, and (c) incurs negligible processing, bandwidth and processing overheads compared to non-collaborative baselines.

关键词： Deep learning Runtime Computational modeling machine vision Conferences Pipelines Neural networks

来源：评论

学校读者我要写书评

暂无评论

Development, Calibration, and image processing of Underwater Structured Light vision System: A Survey

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2023年 72卷 1页

作者： Fan, Junfeng Wang, Xuan Zhou, Chao Ou, Yaming Jing, Fengshui Hou, Zengguang Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China China Univ Min & Technol Sch Mech Elect & Informat Engn Dept Mech Engn Beijing 100083 Peoples R China

The underwater structured light vision system has the characteristics of a large amount of information, high measurement accuracy, and good robustness, which has attracted the attention of researchers and is widely used in the underwater 3-D measurement. In this article, the state-of-the-art techniques of the underwater structured light vision system are comprehensively summarized, which provides valuable guidance for relevant researchers. First, underwater structured light vision systems are classified and introduced, and their advantages and disadvantages are analyzed. Then, the calibration methods of the underwater structured light vision system are introduced, including underwater camera calibration, underwater laser calibration, and underwater projector calibration. Besides, the image processing methods of the underwater structured light vision system are analyzed, including laser structured light image processing and coded structured light image processing. Moreover, the applications of the underwater structured light vision system in underwater archeology, seabed mapping, and underwater structure inspection are introduced. On this basis, the research directions of the underwater structured light vision system are prospected.

关键词： machine vision Measurement by laser beam Three-dimensional displays Cameras Sea measurements Calibration image processing image processing system calibration system development underwater structured light vision system

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Median Filter Hardware Chip for image and Signal processing

Performance Evaluation of Median Filter Hardware Chip for Im...

引用

Cognitive Computing in Engineering, Communications, Sciences and Biomedical Health Informatics (IC3ECSBHI), International Conference on

作者： Sreesh Gaur Akash Goel Pawan Kumar Pal Amit Kumar Singh Sanger Department of Computer Science KIET Group of Institutions Ghaziabad India

ISBN: (数字)9798331518523

ISBN: (纸本)9798331518530

The median filter is a valuable image processing tool that can be used in many applications to advance picture quality, and inferior noise, and acquire data ready for more analysis. The median filter, a non-linear image processing technique, has excessive application in many diverse fields because of its capacity to remove noise while preserving edges. The filter is significant for both simple and complex imageprocessing occupations since it is together easy to comprehend and fast to use. An active method to attain real-time performance is to devise a median filter on a FieldProgrammable Gate Array (FPGA) for image signal processing and machine vision applications. FPGAs are well-suited for photo filtering due to their capability to achieve parallel processing, which is one of their prominent advantages. The research article meets with the design and simulation of the average filter in HDL and synthesis on the FPGA for assessing the performance indices.

关键词： Performance evaluation Filtering machine vision Noise Parallel processing Logic gates Hardware Real-time systems Delays Field programmable gate arrays

来源：评论

学校读者我要写书评

暂无评论

A machine learning framework for the identification of crops and weeds based on shape curvature and texture properties

引用

International Journal of Information Technology (Singapore) 2024年第2期16卷 1261-1274页

作者： Agarwal, Diwakar Electronics and Communication Engineering GLA University Mathura India

Identification of weeds in a crop is a challenging problem for designing an automotive weed management system. Due to similar and indistinguishable properties between crop and weed, rather than single, various type of information is required for the successful discrimination of weeds against crops. This paper presents a machine learning-enabled weed classification system based on the fusion of a plant’s handcrafted shape and texture features at the feature level. The shape features include several shape factors, region-based (Hu moment invariants), and contour-based descriptors (Fourier descriptors and shape curvature). Texture features include GLCM and LBP. A series of experiments are conducted to investigate the utility of shape curvature in the classification of weeds against crops. Furthermore, the potential of shape features including shape curvature and texture features has been examined by modeling the four classifiers, namely, SVM, KNN, MLP, and Naïve Bayes by using the 10-fold cross-validation approach. The classifiers are compared based on performance metrics- accuracy, error rate, recall, specificity, precision, and F1-score. Experimental results show that the SVM classifier achieved 93.67% overall accuracy with shape curvature features in the classification between crop and weed. In addition, the SVM classifier achieved 99.33% overall accuracy with the integration of shape with shape curvature and texture features. © The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer applications and Management 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s);author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

关键词： Classifier Computer vision Crop image processing machine learning Weed

来源：评论

学校读者我要写书评

暂无评论

Computer-Aided Diagnosis System for Lung Fibrosis: From the Effect of Radiomic Features and Multi-Layer-Perceptron Mixers to Pre-Clinical Evaluation

引用

IEEE ACCESS 2024年 12卷 25642-25656页

作者： Fontanellaz, M. Christe, A. Christodoulidis, S. Dack, E. Roos, J. Drakopoulos, D. Sieron, D. Peters, A. Geiser, T. Funke-Chambour, M. Heverhagen, J. Hoppe, H. Exadaktylos, A. K. Ebner, L. Mougiakakou, S. Univ Bern ARTORG Ctr Biomed Engn Res AI Hlth & Nutr CH-3012 Bern Switzerland Univ Bern Univ Hosp Bern Dept Diagnost Intervent & Pediat Radiol Inselspital CH-3012 Bern Switzerland Univ Paris Saclay Lab Math & Informat MICS CentraleSupelec F-91190 Gif Sur Yvette France Luzerner Kantonsspital Inst Radiol & Nucl Med CH-6000 Luzern Canton Of Lucer Switzerland Silesian Ctr Heart Dis Div Magnet Resonance Imaging PL-41800 Zabrze Poland Univ Bern Univ Hosp Bern Dept Pulm Med Inselspital CH-3012 Bern Switzerland Lindenhofspital Dept Radiol CH-3001 Bern Switzerland Campus Stiftung Lindenhof CH-8090 Bern Switzerland Swiss Inst Transnatl & Entrepreneurial Med CH-3010 Bern Switzerland Univ Bern CH-3012 Bern Switzerland Univ Lucerne Dept Hlth Sci & Med CH-6002 Luzern Switzerland Univ Bern Univ Hosp Bern Dept Emergency Med Inselspital CH-3012 Bern Switzerland Cantonal Hosp Lucerne Dept Radiol CH-6000 Luzern Switzerland

Medical image segmentation is a crucial element of computer-aided diagnosis (CAD) systems. Segmentation maps are used to calculate imaging features, such as quantitative disease distribution and radiomic features. Since their introduction in 2015, UNets have become the state-of-the-art segmentation tools. However, since that time, many new methods for image processing have been introduced, such as vision transformers and multi-layer-perceptron-mixers (MLP-Mixers). Alongside baseline UNets, we have now investigated the application of such MLP-Mixers for medical image segmentation, as part of a CAD system for the diagnosis of interstitial lung diseases (ILDs). Furthermore, we have investigated the effect of 2D and 3D data representations on segmentation and the final CAD results. We have evaluated the performance of the baseline segmentation methods and the MLP-Mixer primary on the overall diagnostic performance of the CAD system - as well as on the accuracy of segmentation as an intermediate step. In addition to network and data representation variations, we have investigated two different techniques for selecting features, an agnostic method and an alternative approach which selects features tailored to a specific segmentation map and diagnosis task. Finally, the CAD's performance was compared with that of four independent specialists in chest radiology. Among the 105 test cases, the diagnostic accuracy was 77.2 +/- 1.6% for the AI-approaches and 79.0 +/- 6.9% for the radiologists, indicating that the proposed systems perform comparably well to human readers in most of the cases. For the task of ILD pattern segmentation, similar results were obtained with 3D data and 2D tomography slices.

关键词： image segmentation Computed tomography Pulmonary diseases Databases Lung Task analysis Imaging X-ray applications Chest X-ray chest computed tomography interstitial lung diseases radiomics computer-aided diagnosis segmentation machine learning

来源：评论

学校读者我要写书评

暂无评论

Retinal image Analysis for Detection of Diabetic Macular Edema Using Transfer Learning 8

Retinal Image Analysis for Detection of Diabetic Macular Ede...

引用

8th International Conference on Data Science and machine Learning applications, CDMA 2025

作者： Asim, Tayyaba Akbar, Shahzad Shahzore, Usama Rehman, Amjad Saba, Tanzila Ayesha, Noor Riphah College of Computing Riphah International University Faisalabad Pakistan Prince Sultan University AIDA Lab. CCIS Riyadh Saudi Arabia Zhengzhou University Zhengzhou China Henan450001 China

ISBN: (纸本)9798331539696

Diabetic Macular Edema (DME) is a disease of the eye's retina and it's a major factor of causing vision problems and leads towards blindness if it is undiagnosed. Early detection of DME can prevent vision loss and may reduce diabetic-related problems like cardiovascular issues. Therefore, this study presents a method for detecting DME through Optical Coherence Tomography (OCT) and fundus images using transfer learning. The proposed method is based on four stages: Pre-processing, augmentation, segmentation through DeepLabV3+, and binary classification of DME by using two (02) publicly available datasets;the first dataset is Messidor-2, which contains fundus images, and the second dataset is Retinal images of Optical Coherence Tomography (OCT). In the Messidor-2 dataset, the total number of images is 1744, and Retinal OCT images dataset consists of 84,495 images total, separated into four categories (CNV, DME, DRUSEN, NORMAL). In the proposed method, Convolutional Neural Networks (CNN) architectures ResNet50 and VGG-19 have been used for the detection of the DME. Convolutional Neural Networks (CNNs) have been extensively used in medical imaging analysis and classification. Using the well-known ResNet50 architecture for classification of each dataset, the proposed model yielded an accuracy of 98.79%, 99% of F1 Score, 98.43% of Precision, and recall of 98.89%. By using VGG-19, the proposed model gives an accuracy of 98.81%, 98.94% of Fl Score, 98.1 % of Precision and recall of 98.73%. When both models (ResNet50 and VGG-19) were compared, the VGG-19 gave the best accuracy. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Computer-vision-Based Industrial Algorithm for Detecting Fruit and Vegetable Dimensions and Positioning 19th

Computer-Vision-Based Industrial Algorithm for Detecting Fr...

引用

19th International Conference on Soft Computing Models in Industrial and Environmental applications, SOCO 2024

作者： Guevara, Cesar Rostan, Julen Rodriguez, Javier Gonzalez, Silvia Sedano, Javier C. López Bravo 70 Burgos09001 Spain

ISBN: (纸本)9783031750090

A computer-vision-based industrial algorithm is proposed in this study for the detection of the dimensions and the spatial positioning of fruit and vegetables on a conveyor belt for their movement to a packing machine with a robotic arm. The principal purpose of the algorithm is to identify the dimensions (length, width), and position (angle of inclination, and Cartesian coordinates) of the vegetable mesh netting without taking the product label into account. The proposed algorithm has four functions. A convolutive neuronal network model is applied for object detection, with which all objects are identified in the image while the product label is suppressed, so that recognition of the vegetable mesh netting and its dimensions is not impaired. Moreover, the Canny edge detector and the Border following algorithms are applied to perform image pre-processing and edge improvements, respectively, yielding optimal results with more clearly defined noise-free objects. The feature extraction function for dimensions yielded detection precision results of 99.60% for length, 99.00% for width, and 97.80% for inclination angle, showing optimal performance in the test phases. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Multi-Sensor Surveillance System Based on Integrated Video Analytics

引用

IEEE SENSORS JOURNAL 2022年第11期22卷 10207-10222页

作者： Purohit, Manoj Singh, Manvendra Yadav, Saralesh Singh, Adarsh Kumar Kumar, Ajay Kaushik, Brajesh Kumar Instruments Res & Dev Estab Dehra Dun 248008 Uttarakhand India Indian Inst Technol Roorkee Dept Elect & Commun Engn Roorkee 247667 Uttarakhand India

image-based sensing is a powerful source of information for a wide range of applications such as surveillance, monitoring, and automation. The volume of data collected by surveillance devices is enormous;therefore, images are processed to extract application-driven data, moreover, centralized processing is unfeasible when several cameras are distributed across broad geographical regions. In such scenario, with processing limitations and bandwidth availability being important restrictions, an automated process is required for extracting meaningful information from the collected data during surveillance process and reducing operator fatigue. The proposed design, which continuously processes the image stream to meet real-time requirements, provides integrated video analytics with enhancement, panoramic views, and automated motion detection. This system enhances pictorial information for human interpretation and analyzes scene data using machine-vision modules. These features aid in the use of surveillance system for continuous monitoring of border areas without the need for human intervention under adverse-weather conditions. The proposed system architecture is based on distributed video-analytics using visual and infrared-band sensors. The system has a visual detection range of 5 km with 80% detection probability for a 2.3 m X 2.3 m target under standard clear visibility conditions. The preliminary results indicate that the system is capable of processing up to 25 video frames per second. Additionally, the implemented system records data for post-analysis and uses video analytics to reduce the bandwidth requirement of communication link. Finally, this system can be distinguished because it is flexible, compact, affordable, and scalable, and therefore, can be considered for monitoring and surveillance.

关键词： Surveillance Sensors Sensor systems Cameras Visual analytics Security Feature extraction Distributed architecture image enhancement intelligent video system motion detection real-time video analytics surveillance system

来源：评论

学校读者我要写书评

暂无评论

CapNet: An Encoder-Decoder based Neural Network Model for Automatic Bangla image Caption Generation

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND applications 2022年第8期13卷 752-759页

作者： Rahman, Rashik Saha, Aloke Kumar Murad, Hasan Al Masud, Shah Murtaza Rashid Rahman, Nakiba Nuren Momtaz, A. S. Zaforullah Univ Asia Pacific Comp Sci & Engn Dhaka Bangladesh Chittagong Univ Engn & Technol Comp Sci & Engn Chattogram Bangladesh

Automatic caption generation from images has become an active research topic in the field of Computer vision (CV) and Natural Language processing (NLP). machine generated image caption plays a vital role for the visually impaired people by converting the caption to speech to have a better understanding of their surrounding. Though significant amount of research has been conducted for automatic caption generation in other languages, far too little effort has been devoted to Bangla image caption generation. In this paper, we propose an encoder-decoder based model which takes an image as input and generates the corresponding Bangla caption as output. The encoder network consists of a pretrained image feature extractor called ResNet-50, while the decoder network consists of Bidirectional LSTMs for caption generation. The model has been trained and evaluated using a Bangla image captioning dataset named BanglaLekhaimageCaptions. The proposed model achieved a training accuracy of 91% and BLEU-1, BLEU-2, BLEU-3, BLEU-4 scores of 0.81, 0.67, 0.57, and 0.51 respectively. Moreover, a comparative study for different pretrained feature extractors such as VGG-16 and Xception is presented. Finally, the proposed model has been deployed on an embedded device for analysing the inference time and power consumption.

关键词： -Bangla image caption generation encoder-decoder bidirectional long short term memory (LSTM) bangla natural language processing (NLP)

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：