检索结果-内蒙古大学图书馆

Using an LCD Monitor and a Robotic Arm to Quickly Establish image Datasets for Object Detection

IEEE ACCESS 2021年 9卷 131006-131019页

作者： Lee, Wei-Chen Zhang, Jen-You Wei, Ching-Chih Natl Taiwan Univ Sci & Technol Dept Mech Engn Taipei 106335 Taiwan Natl Taiwan Univ Sci & Technol Ctr Cyber Phys Innovat Taipei 106335 Taiwan MPI Corp Zhubei City 302047 Taiwan HSL Machinery Works Co Ltd Tainan 717020 Taiwan

This study proposes a method for quickly establishing a large image dataset required for object detection. One way of obtaining a large image dataset is to use the green screen capturing method to collect sample images. The limitation is that the color of the object to be photographed must be different from the green background. To solve this issue, this study used an LCD monitor that can change the background color to capture the object images, and it then adopted image processing techniques to obtain the sample images and their masks for image synthesis. This study also used a robotic arm to move the camera to different locations to capture the images from various view directions to increase the diversity of the sample images. A Faster R-CNN model and a YOLOv4 model were used as the deep learning models. After training the models with 10 000 synthetic images that were created by using the proposed method, we could achieve an average mAP of 0.76 for the Faster R-CNN model and that of 0.97 for the YOLOv4 model, which were higher than the average mAP of 0.66 for the Faster R-CNN model and of 0.93 for the YOLOv4 model, both trained with 800 real images. Although more synthetic images were required to achieve better accuracy, only about 2 h were spent to prepare the 10 000 annotated synthetic images, saving about 69% of the required time to capture and label 800 real images. Another advantage of the method is that it generates instance segmentation annotations, which can be easily converted to other annotation types.

关键词： image color analysis Object detection image segmentation Training Liquid crystal displays Webcams Solid modeling LCD monitor robotic arm object detection synthetic image faster R-CNN YOLO

来源：评论

学校读者我要写书评

暂无评论

ONNS: Experimental Evaluation of Human Smile Recognition Based on image processing Assisted Optimized Neural Network Strategy

ONNS: Experimental Evaluation of Human Smile Recognition Bas...

引用

Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES), International Conference on

作者： K. Murugan V. Srilakshmi A. Swathi Sunheriya Neeraj Suha Afaneh C. Thirupathaiah Department of ECE Bannari Amman Institute of Technology Sathvamangalam Department of S&H QIS College of Engineering and Technology Prakasam District AndhraPradesh India Department of CSE QIS College of Engineering and Technology Prakasam District AndhraPradesh India Department of Mechanical Engineering Yeshwantrao Chavan College of Engineering Nagpur Department of cybersecurity faculty of information technology Zarqa university Zarqa Jordan University of Business and Technology Jeddah Saudi Arabia Department of Mechanical Engineering Annamacharya University Rajampet

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543631

This paper presents the experimental evaluation of human smile recognition using an image processing-assisted optimized neural network strategy (ONNS). The primary objective was to assess different neural network architectures and feature extraction techniques to improve the accuracy and real-time performance of smile detection. Several deep learning architectures, including a standard CNN, optimized CNN, BKNet, RetinaNet, and MobileNet, were evaluated on a smile recognition dataset. Geometric and texture-based feature extraction techniques such as Local Binary Patterns (LBP) and Histogram of Oriented Gradients (HOG) were employed to improve the robustness of the model. The results showed that the optimized CNN achieved the highest accuracy of 95.3%, with a precision of 93.6% and a recall of 94.1%. Furthermore, a real-time performance evaluation demonstrated that MobileNet, while achieving the fastest inference time of 10 ms (100 fps), had a slightly lower accuracy of 90.5%. The combination of geometric and texture features led to improved performance, with the combined method achieving an accuracy of 92%. These results indicate that the proposed optimized neural network strategy is effective in improving smile recognition accuracy while maintaining real-time performance.

关键词： Performance evaluation Histograms Accuracy image recognition Neural networks Feature extraction real-time systems Robustness Convolutional neural networks Standards

来源：评论

学校读者我要写书评

暂无评论

A Quasi-conformal Mapping-based Data Augmentation Technique For Brain Tumor Segmentation

A Quasi-conformal Mapping-based Data Augmentation Technique ...

引用

Medical Imaging Conference - image processing

作者： Zhang, Min An, Dongsheng Young, Geoffrey S. Gu, Xianfeng Xu, Xiaoyin Harvard Med Sch Brigham & Womens Hosp Dept Radiol Boston MA 02115 USA SUNY Stony Brook Dept Comp Sci Stony Brook NY 11794 USA

ISBN: (纸本)9781510633940

deep neural networks (DNNs) have been widely used in the medical imaging field. The large and high quality dataset is crucial for the performance of the deep learning models, but the medical data and ground-truth is often insufficient and very expensive in terms of time and human effort on the data collection. However, we can improve the performance of the deep learning model by augmenting the data we already have. In this work, we introduce a novel differential geometry-based quasi conformal (QC) mapping augmentation technique to augment the brain tumor images. The QC method lets the user specify or randomly generate a complex-valued function on the image domain via Beltrami coefficient. By solving the Beltrami equation with given Beltrami coefficient, the QC map, which can further guide the deformation of the image, is able to generate all possible linear and non-linear image warpings and it is flexible to allow the user to fully control the global and local deformations. Our experimental results demonstrate the efficiency and efficacy of the proposed method.

关键词： data augmentation Beltrami equation quasi-conformal mappings brain tumor segmentation deep learning

来源：评论

学校读者我要写书评

暂无评论

A novel difficult-to-segment samples focusing network for oral CBCT image segmentation

引用

SCIENTIFIC REPORTS 2024年第1期14卷 1-9页

作者： Hu, Fengjun Chen, Zeyu Wu, Fan Zhejiang Shuren Univ Coll Informat Sci & Technol Hangzhou 310015 Peoples R China Zhejiang Shuren Univ Zhejiang Netherlands Joint Lab Digital Diag & Trea Hangzhou 310015 Zhejiang Peoples R China

Using deep learning technology to segment oral CBCT images for clinical diagnosis and treatment is one of the important research directions in the field of clinical dentistry. However, the blurred contour and the scale difference limit the segmentation accuracy of the crown edge and the root part of the current methods, making these regions become difficult-to-segment samples in the oral CBCT segmentation task. Aiming at the above problems, this work proposed a Difficult-to-Segment Focus Network (DSFNet) for segmenting oral CBCT images. The network utilizes a Feature Capturing Module (FCM) to efficiently capture local and long-range features, enhancing the feature extraction performance. Additionally, a Multi-Scale Feature Fusion Module (MFFM) is employed to merge multiscale feature information. To further improve the loss ratio for difficult-to-segment samples, a hybrid loss function is proposed, combining Focal Loss and Dice Loss. By utilizing the hybrid loss function, DSFNet achieves 91.85% Dice Similarity Coefficient (DSC) and 0.216 mm Average Symmetric Surface Distance (ASSD) performance in oral CBCT segmentation tasks. Experimental results show that the proposed method is superior to current dental CBCT image segmentation techniques and has real-world applicability.

关键词： deep learning images segmentation Medical image processing Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Multi-year Crop Type Mapping using pre-trained deep Long-Short Term Memory and Sentinel 2 image time Series 27

Multi-year Crop Type Mapping using pre-trained deep Long-Sho...

引用

Conference on image and Signal processing for Remote Sensing XXVII

作者： Weikmann, Giulio Paris, Claudia Bruzzone, Lorenzo Univ Trento Dept Informat Engn & Comp Sci Via Sommar 5 I-38123 Trento Italy

ISBN: (数字)9781510645691

ISBN: (纸本)9781510645691;9781510645684

This work presents a system for multi-year crop type mapping based on the multi-temporal Long Short-Term Memory (LSTM) deep learning (DL) model and Sentinel 2 image time Series (TS). The method assumes the availability of a pre-trained LSTM model for a given year and aims to update the corresponding crop type map for a different year considering a small amount of recent reference data. To this end, the proposed approach combines Self-Paced learning (SPL) and fine-tuning (FT) techniques. While the SPL technique gradually incorporates samples from crop types that can be classified with high-confidence by the pre-trained model, the FT strategy adapts the network to those classes having low-confidence accuracy. This condition allows us to reduce the labeled samples required to achieve accurate classification results. The experimental results obtained on three tiles of the Austrian country on TSs of Sentinel 2 data acquired in 2019 and 2020 (considering a model pre-trained on images of 2018) demonstrate the capability of the LSTM to adapt to TS of images with different temporal and radiometric characteristic with respect to the one used to pre-train the model, with a relatively small number of training samples. As expected, by directly applying the model without performing any adaptation, we obtain a mean F-score (F1%) of 64% and 62% compared to 76% and 70% achieved by the proposed technique with only 1500 samples for 2019 and 2020, respectively.

关键词： Long Short Term Memory (LSTM) automatic classification multi-temporal deep learning (DL) multi-temporal analysis multi-year crop type mapping remote sensing

来源：评论

学校读者我要写书评

暂无评论

Agricultural Pest and Disease Detection System using Machine learning

Agricultural Pest and Disease Detection System using Machine...

引用

Green Industrial Electronics and Sustainable Technologies (GIEST), IEEE International Conference on

作者： Boyai Salam Khundrakpam Johnson Singh Computer Science and Engineering National Institute of Technology Imphal Manipur India

ISBN: (数字)9798331540142

ISBN: (纸本)9798331540159

Crop diseases pose a significant threat to agricultural productivity in Manipur, a state renowned for its rich agricultural heritage and diverse crop cultivation practices. Early detection of these diseases is crucial for safeguarding the region's agricultural sustainability and ensuring food security. In this paper, we present a comprehensive machine learning-based system utilizing Convolutional Neural Network (CNN) and the U-NET model for detecting crop diseases specifically tailored to the unique agricultural landscape of Manipur, India. Leveraging advanced techniques such as transfer learning with deep learning pre-trained models, including VGG16 and VGG19, we developed a robust system capable of accurately identifying diseased crops, including locally grown vegetables like "Kobiphun" (cabbage), "Kobilei" (cauliflower), "Aloo" (potato), "Khamen Asinba" (tomato), and "Chak-hao" (black rice). The datasets used in this project are purely self-collected primary data, reflecting extensive effort and dedication in gathering high-quality images for training and model development. Our methodology involves detailed image processing, feature extraction demonstrating the effectiveness of our approach in addressing the specific challenges posed by crop diseases in Manipur's agricultural system. The findings highlight the model's high accuracy (over 92%), precision (exceeding 90%), and recall (approaching 100%), indicating its reliability in real-world applications. Additionally, the system's scalability allows for integration with real-time monitoring tools, enhancing its applicability for local farmers. This research contributes to improved disease management practices, enhances the resilience of Manipur's agriculture sector, and ensures the livelihoods of its farming communities.

关键词： deep learning Visualization Accuracy Transfer learning Crops Food security real-time systems Convolutional neural networks Sustainable development Diseases

来源：评论

学校读者我要写书评

暂无评论

深度学习图像合成研究综述

引用

中国图象图形学报 2023年第12期28卷 3670-3698页

作者：叶国升王建明杨自忠张宇航崔荣凯宣帅大理大学数学与计算机学院大理671003 云南省昆虫生物医药研发重点实验室(大理大学) 大理671000

图像合成一直是图像处理领域的研究热点,具有广泛的应用前景。从原图中精确提取出前景目标对象并将其与新背景合成,构造尽量接近真实的图像是图像合成的基本目标。为推动基于深度学习的图像合成技术研究与发展,本文论述了当前图像合成... 详细信息

图像合成一直是图像处理领域的研究热点,具有广泛的应用前景。从原图中精确提取出前景目标对象并将其与新背景合成,构造尽量接近真实的图像是图像合成的基本目标。为推动基于深度学习的图像合成技术研究与发展,本文论述了当前图像合成任务中面临的主要问题:(1)前景对象适应性问题,包括前景对象相对于背景图像的大小、位置、几何角度等几何一致性问题,以及前后景互相遮挡、前景对象边缘细节模糊的外观一致性问题;(2)视觉和谐问题,包括前后景色彩、对比度、饱和度等不统一的色调一致性问题,及前景对象丢失对应阴影的阴影缺失问题;(3)生境适应性问题,表现为前景对象与背景图像的逻辑合理性。总结了目前为解决不同问题主要使用的深度学习方法,同时对不同问题中的合成图像结果进行质量评估,总结了相应的评价指标,并介绍了为解决不同问题所使用的公开数据集,同时进行了深度学习方法的对比,描述了图像合成技术的主要应用场景,最后分析了基于深度学习的图像合成技术中仍然存在的不足,同时提出可行的研究意见,并对未来图像合成技术发展方向提出展望。

关键词：深度学习图像合成前景对象适应性图像和谐化生境适应性

来源：评论

学校读者我要写书评

暂无评论

Integrating Spatial and Temporal Information for Violent Activity Detection from Video Using deep Spiking Neural Networks

引用

SENSORS 2023年第9期23卷 4532页

作者： Wang, Xiang Yang, Jie Kasabov, Nikola K. Shanghai Jiao Tong Univ Inst Image Proc & Pattern Recognit Shanghai 200400 Peoples R China Auckland Univ Technol Knowledge Engn & Discovery Res Inst Auckland 1020 New Zealand

Increasing violence in workplaces such as hospitals seriously challenges public safety. However, it is time- and labor-consuming to visually monitor masses of video data in real time. Therefore, automatic and timely violent activity detection from videos is vital, especially for small monitoring systems. This paper proposes a two-stream deep learning architecture for video violent activity detection named SpikeConvFlowNet. First, RGB frames and their optical flow data are used as inputs for each stream to extract the spatiotemporal features of videos. After that, the spatiotemporal features from the two streams are concatenated and fed to the classifier for the final decision. Each stream utilizes a supervised neural network consisting of multiple convolutional spiking and pooling layers. Convolutional layers are used to extract high-quality spatial features within frames, and spiking neurons can efficiently extract temporal features across frames by remembering historical information. The spiking neuron-based optical flow can strengthen the capability of extracting critical motion information. This method combines their advantages to enhance the performance and efficiency for recognizing violent actions. The experimental results on public datasets demonstrate that, compared with the latest methods, this approach greatly reduces parameters and achieves higher inference efficiency with limited accuracy loss. It is a potential solution for applications in embedded devices that provide low computing power but require fast processing speeds.

关键词： violence detection deep learning spiking neural network optical flow spatial and temporal analysis

来源：评论

学校读者我要写书评

暂无评论

Pest Identification and Control using deep learning and Augmented reality

Pest Identification and Control using Deep Learning and Augm...

引用

2022 International Conference on Edge Computing and Applications, ICECAA 2022

作者： Soni, Ascharya Verma, Sachin Khare, Anuraag Asha Rani, K.P. Darshan Balaji, P.S. Gowrishankar, S. Department of Computer Science and Engineering Dr. Ambedkar Institute of Technology Karnataka Bengaluru560056 India

ISBN: (数字)9781665482325

ISBN: (纸本)9781665482325

It is crucial to comprehend how insect pest populations affect the subsequent yield or harvest since the ultimate goal of agriculture is to provide a sustained economic production of crop products. Using pesticides is the simplest technique to manage the pest infestation. However, using pesticides improperly or in excess can harm both people and animals as well as the plants. Machine learning algorithms and image processing techniques are widely used in agricultural research, and they have significant potential, particularly in plant protection, which ultimately leads to crop management. This paper highlights the detection of pests and their control measures. A smartphone camera will capture photographs of the pests through a mobile app built using the Flutter framework. The images are then analyzed in the app using various transfer learning based models for available pest identification kaggle dataset. The flutter framework offers the ability to monitor targets in real-time on a mobile device and aids in visualizing the detected pest by integrating augmented reality on to the app. © 2022 IEEE.

关键词： Crops

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE 30th International Symposium on Field-Programmable Custom Computing Machines (FCCM)

作者： Park, Yeo-Reum Kim, Ji-Hoon Do, Jaeyoung Kim, Joo-Young Korea Adv Inst Sci & Technol Sch Elect Engn Daegeon South Korea Amazon Alexa AI Kirkland WA 98033 USA

ISBN: (纸本)9781665483322

image-text retrieval (ITR) that identifies the relevant images for a given text query, or vice versa, is the fundamental task in emerging vision-and-language machine learning applications. Recently, the cross-modal approach that extracts image and text features in separate reasoning pipelines but performs the similarity search on the same embedding representation is proposed for the real-time ITR system. However, the similarity search that finds the most relevant data in huge data embeddings for a given query becomes the bottleneck of the ITR system. In this paper, we propose a dual-mode similarity search accelerator that can solve the computational hurdle for online image-to-text and text-to-image retrieval service. We propose an embedding compression scheme that removes the sparsity in the text embeddings, further eliminating the time-consuming masking operations in the later processing pipeline. Combining with the data quantization from 32-bit floating-point to 8bit integer, we reduce the target dataset size by 95.1% with less than 0.1% accuracy loss for 1024-dimensional embedding features. In addition, we propose a streamlined similarity search data flow for both query types, which minimizes the required memory bandwidth with maximal data reuse. The query and data embeddings are guaranteed to be fetched only once from the external memory with the optimized data flow. Based on the proposed data representation and flow, we design a scalable similarity search accelerator that includes multiple ITR kernels. Each ITR kernel has modular design, composed of a separate memory access module and a computing module. The computing module supports pipelined operations of the four similarity search tasks: dot product calculation, data reordering, partial score aggregation, and ranking. We double the number of processing operations in the computing module with the DSP packing technique. Finally, we implement the proposed accelerator with six ITR kernels on the Xilinx Alveo U28

关键词： image coding Pipelines Graphics processing units Feature extraction Throughput Energy efficiency real-time systems

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：