检索结果-内蒙古大学图书馆

CacheTrack-YOLO: real-time Detection and Tracking for Thyroid Nodules and Surrounding Tissues in Ultrasound Videos

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2021年第10期25卷 3812-3823页

作者： Wu, Xiangqiong Tan, Guanghua Zhu, Ningbo Chen, Zhilun Yang, Yan Wen, Huaxuan Li, Kenli Hunan Univ Coll Comp Sci & Elect Engn Changsha 410082 Peoples R China Wenzhou Med Univ Affiliated Hosp 2 Dept Ultrason Diag Wenzhou 325088 Peoples R China Wenzhou Med Univ Yuying Childrens Hosp Wenzhou 325088 Peoples R China Southern Med Univ sity Affiliated Shenzhen Maternal & Child Healthcare Dept Ultrasound Shenzhen 518047 Peoples R China

To accurately detect and track the thyroid nodules in a video is a crucial step in the thyroid screening for identification of benign and malignant nodules in computer-aided diagnosis (CAD) systems. Most existing methods just perform excellent on static frames selected manually from ultrasound videos. However, manual acquisition is labor-intensive work. To make the thyroid screening process in a more natural way with less labor operations, we develop a well-designed framework suitable for practical applications for thyroid nodule detection in ultrasound videos. Particularly, in order to make full use of the characteristics of thyroid videos, we propose a novel post-processing approach, called Cache-Track, which exploits the contextual relation among video frames to propagate the detection results into adjacent frames to refine the detection results. Additionally, our method can not only detect and count thyroid nodules, but also track and monitor surrounding tissues, which can greatly reduce the labor work and achieve computer-aided diagnosis. Experimental results show that our method performs better in balancing accuracy and speed.

关键词： Thyroid Ultrasonic imaging Videos Solid modeling Feature extraction image segmentation Cancer deep learning thyroid nodules computer-aided diagnosis ultrasound videos detection and tracking

来源：评论

学校读者我要写书评

暂无评论

基于改进YOLOv8n的轻量化马铃薯表面缺陷在线检测方法

引用

农业工程学报 2025年第5期41卷 135-144页

作者：许英超刘书玮王相友吴海涛黄杰王恒仁王毅山东理工大学农业工程与食品科学学院淄博255000

马铃薯表面缺陷是农产品分级的重要依据。为提升马铃薯表面缺陷检测精度并实现模型在移动端的快速识别,该研究提出了一种基于改进YOLOv8n的马铃薯表面缺陷检测方法DATW-YOLOv8。算法使用Dilation-wise Residual模块替换C2f中的Bottlenec... 详细信息

马铃薯表面缺陷是农产品分级的重要依据。为提升马铃薯表面缺陷检测精度并实现模型在移动端的快速识别,该研究提出了一种基于改进YOLOv8n的马铃薯表面缺陷检测方法DATW-YOLOv8。算法使用Dilation-wise Residual模块替换C2f中的Bottleneck模块,并引入Dilated Reparam Block模块对C2f进行二次改进,加强细节特征提取,提高缺陷特征的提取精度;随后,引入轻量级自适应下采样(ADOWN)卷积模块,实现图像数据的有效降维,提升模型处理效率;此外,改造检测头为任务对齐动态检测头(task align dynamic detection head,TADDH),提高缺陷边界预测精度,精准聚焦缺陷关键区;最终,使用Wise-EIoU作为边界框回归损失函数,增强模型对边界模糊样本的关注度,提升缺陷边界回归精度及模型鲁棒性。试验结果表明,改进DATW-YOLOv8模型在准确率、召回率和平均精度方面分别达到95.8%、88.1%和94.3%,参数量和权重分别为1.5 M和3.6 MB。与原YOLOv8n模型相比,参数量和权重分别减少了50.0%和42.9%,同时准确率、召回率和平均精度分别提高了2.8、1.6和1.4个百分点。该方法能满足实际生产中针对缺陷马铃薯进行精准、实时检测的要求,为马铃薯表面缺陷在线检测及模型在移动端的部署提供了技术参考。

关键词：图像识别深度学习无损检测马铃薯表面缺陷轻量化 YOLOv8

来源：评论

学校读者我要写书评

暂无评论

Intelligent image-based colourimetric tests using machine learning framework for lateral flow assays

引用

EXPERT SYSTEMS WITH APPLICATIONS 2020年 139卷 112843-000页

作者： Tania, Marzia Hoque Lwin, Khin T. Shabut, Antesar M. Najlah, Mohammad Chin, Jeannette Hossain, M. A. Anglia Ruskin Univ Fac Sci & Engn Med Technol Res Ctr Chelmsford Essex England Anglia Ruskin Univ Fac Hlth Educ Med & Social Care Chelmsford Essex England Teesside Univ Sch Comp & Digital Technol Middlesbrough Cleveland England Leeds Trinity Univ Sch Arts & Commun Leeds W Yorkshire England Anglia Ruskin Univ Sch Allied Hlth Chelmsford Essex England Univ East Anglia Sch Comp Sci Norwich Norfolk England

This paper aims to deliberately examine the scope of an intelligent colourimetric test that fulfils ASSURED criteria (Affordable, Sensitive, Specific, User-friendly, Rapid and robust, Equipment-free, and Deliverable) and demonstrate the claim as well. This paper presents an investigation into an intelligent image-based system to perform automatic paper-based colourimetric tests in real-time to provide a proof-of-concept for a dry-chemical based or microfluidic, stable and semi-quantitative assay using a larger dataset with diverse conditions. The universal pH indicator papers were utilised as a case study. Unlike the works done in the literature, this work performs multiclass colourimetric tests using histogram-based image processing and machine learning algorithm without any user intervention. The proposed image processing framework is based on colour channel separation, global thresholding, morphological operation and object detection. We have also deployed aserver-based convolutional neural network framework for image classification using inductive transfer learning on a mobile platform. The results obtained by both traditional machine learning and pre-trained model-based deep learning were critically analysed with the set evaluation criteria (ASSURED criteria). The features were optimised using univariate analysis and exploratory data analysis to improve the performance. The image processing algorithm showed >98% accuracy while the classification accuracy by Least Squares Support Vector Machine (LS-SVM) was 100%. On the other hand, the deep learning technique provided >86% accuracy, which could be further improved with a large amount of data. The k-fold cross-validated LS-SVM based final system, examined on different datasets, confirmed the robustness and reliability of the presented approach, which was further validated using statistical analysis. The understaffed and resource-limited healthcare system can benefit from such an easy-to-use technology to support

关键词： image processing Histogram thresholding Feature selection Machine learning deep learning Colourimetric tests

来源：评论

学校读者我要写书评

暂无评论

Efficient Hardware Design of Convolutional Neural Networks for Accelerated deep learning

Efficient Hardware Design of Convolutional Neural Networks f...

引用

Midwest Symposium on Circuits and Systems (MWSCAS)

作者： Kasem Khalil Md Rahat Khan Magdy Bayoumi Ahmed Sherif Department of Electrical and Computer Engineering University of Mississippi Mississippi USA Department of Electrical and Computer Engineering University of Louisiana Lafayette LA USA School of Computing Sciences and Computer Engineering University of Southern Mississippi MS USA

ISBN: (数字)9798350387179

ISBN: (纸本)9798350387186

The hardware design of Convolutional Neural Networks (CNNs) significantly accelerates deep learning tasks, providing substantial improvements in computational efficiency and real-time processing capabilities. This continuous improvement of hardware design enhances the speed of convolutional operations by optimizing resource utilization and energy efficiency. Consequently, CNNs become more accessible for deployment in resource-constrained environments, facilitating advancements in a wide range of applications, from image recognition to embedded systems. In this paper, we propose a novel hardware design for CNNs demonstrating enhanced efficiency and performance. The proposed method incorporates a shared convolutional unit shared between multiple kernel units. The proposed design leverages a meticulous allocation of hardware resources. The proposed method is implemented on the Aletra 10 GX FPGA using the VHDL language. The CNN accelerator showcases remarkable results, with resource utilization percentages ranging from 2% to 26.35%. Operating at a clock frequency of 120 MHz with an 8-bit fixed design precision, the proposed method achieves superior energy efficiency at 28.41 GOPS/W, outperforming existing approaches. Experimental results show that testing with the CIFAR-10 dataset demonstrates a remarkable accuracy of 98.79%. Additionally, the novel design has a lower occupied area than the existing ones. This hardware design promises significant advancements in CNN acceleration and offers a robust solution for applications demanding both resource efficiency and high accuracy.

关键词： deep learning Power demand Accuracy VHDL Hardware Energy efficiency Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Classification Algorithms for Millimeter-wave Imaging 16

Performance Analysis of Classification Algorithms for Millim...

引用

16th European Conference on Antennas and Propagation (EuCAP)

作者： Sharma, Rahul Hussung, Raphael Keil, Andreas Friederich, Fabian Fromenteze, Thomas Khalily, Mohsen Deka, Bhabesh Fusco, Vincent Yurduseven, Okan Queens Univ Belfast ECIT Belfast BT3 9DT Antrim North Ireland Fraunhofer Inst Ind Math ITWM Dept Mat Characterizat & Testing D-67663 Kaiserslautern Germany Univ Limoges XLIM UMR 7252 F-87000 Limoges France Univ Surrey Inst Commun Syst Guildford GU2 7XH Surrey England Tezpur Univ Dept Elect & Commun Engn Tezpur 784028 Assam India

ISBN: (纸本)9788831299046

A detailed analysis of the performance of traditional machine learning and deep learning techniques applied on a representative classification problem of millimeter-wave (mmW) images is presented in this paper. The algorithms chosen for this analysis are the k-Nearest Neighbors (KNN), Random Forest (RF) and Convolutional Neural Network (CNN). All algorithms presented here are modeled using 'keras' library inside TensorFlow and 'scikit-learn' module. The dataset for training and testing are generated via a developed near-field coded aperture computational imaging (CI) physical model. The use of a physical model of an imaging system that implements CI techniques instead of an experimental set-up makes the whole dataset generation process facile and less time consuming. The training data, in case of the RF and KNN algorithms, are presented in tabular form whereas for the CNN technique, the synthesized images from the physical model itself are used for training. The models are tested with both synthesized as well as experimental data, generated from the physical model and a mmW handheld imager, respectively. Upon testing, it is observed that the KNN and RF algorithms are able to classify the test samples with accuracies of 82% and 87%, respectively, whereas an accuracy of 90% is observed in case of the CNN classifier. Also, an inference speed test is conducted on all the three algorithms. It was observed that CNN is the fastest to predict classes for all of the test samples with a frame rate of 3.8 ms/sample whereas RF is the slowest, with a frame rate of 65.9 ms/sample. These findings establish the fact that when it comes to image classification, CNN based classifiers perform better than any traditional machine learning algorithms with more accurate and faster predictions, paving the way for various real-time applications such as automatic threat detection.

关键词： Convolutional Neural Networks K-nearest neighbours random forest classifiers image classification computational imaging millimeter-waves

来源：评论

学校读者我要写书评

暂无评论

Telemetry Data Compression Algorithm Based on Balanced Cyclic Neural Network and deep learning

Telemetry Data Compression Algorithm Based on Balanced Cycli...

引用

image processing and Computer Applications (ICIPCA), IEEE International Conference on

作者： Min Huang School of Computer Sichuan Technology and Business University Chengdu China

With the rapid development of telemetry and telecontrol technology, the amount of information for acquiring telemetry data is increasing, which requires a large amount of storage space when storing, a large bandwidth when transmitting, and data processing as fast as possible. Therefore, data compression technology is needed to solve this problem. The redundancy of telemetry data is also very large. Statistics show that the redundancy of data transmitted in the standard telemetry system reaches more than 90%. In order to reduce the signal space and relieve the pressure of data processing, it is necessary and important to use data compression technology in telemetry system. Considering the space performance and time performance of the lossless compression algorithm, several commonly used lossless compression algorithms are compared for telemetry noise data. The compression space performance of each algorithm still needs to be improved. Considering that the measured engineering data is distributed in a typical balanced way, we studied the anomaly detection problem of time series data with inter class balance, and proposed an anomaly detection problem based on recurrent neural network (RNN). Finally, the overall test of the developed data acquisition and compression system is carried out. The test results show that the system has completed the real-time acquisition and compression of six channels of rapidly varying parameters, and the compression ratio is more than 1.8:1. All indicators meet the design requirements.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Spectrum-inspired Low-light image Translation for Saliency Detection 22

Spectrum-inspired Low-light Image Translation for Saliency D...

引用

Proceedings of the Thirteenth Indian Conference on Computer Vision, Graphics and image processing

作者： Kitty Varghese Sudarshan Rajagopalan Mohit Lamba Kaushik Mitra Indian institute of technology Madras India Madras Institute of Technology India Indian Institute of Technology Madras India IIT Madras India

ISBN: (纸本)9781450398220

Saliency detection methods are central to several real-world applications such as robot navigation and satellite imagery. However, the performance of existing methods deteriorate under low-light conditions because training datasets mostly comprise of well-lit images. One possible solution is to collect a new dataset for low-light conditions. This involves pixel-level annotations, which is not only tedious and time-consuming but also infeasible if a huge training corpus is required. We propose a technique that performs classical band-pass filtering in the Fourier space to transform well-lit images to low-light images and use them as a proxy for real low-light images. Unlike popular deep learning approaches which require learning thousands of parameters and enormous amounts of training data, the proposed transformation is fast and simple and easy to extend to other tasks such as low-light depth estimation. Our experiments show that the state-of-the-art saliency detection and depth estimation networks trained on our proxy low-light images perform significantly better on real low-light images than networks trained using existing strategies.

关键词： Low-light and salient object detection

来源：评论

学校读者我要写书评

暂无评论

Safety Helmet Detection Based on YOLOv5 Driven by Super-Resolution Reconstruction

引用

SENSORS 2023年第4期23卷 1822页

作者： Han, Ju Liu, Yicheng Li, Zhipeng Liu, Yan Zhan, Bixiong China Construct First Grp Construct & Dev Co Ltd Beijing 100102 Peoples R China Sichuan Univ Coll Elect Engn Chengdu 610065 Peoples R China

High-resolution image transmission is required in safety helmet detection problems in the construction industry, which makes it difficult for existing image detection methods to achieve high-speed detection. To overcome this problem, a novel super-resolution (SR) reconstruction module is designed to improve the resolution of images before the detection module. In the super-resolution reconstruction module, the multichannel attention mechanism module is used to improve the breadth of feature capture. Furthermore, a novel CSP (Cross Stage Partial) module of YOLO (You Only Look Once) v5 is presented to reduce information loss and gradient confusion. Experiments are performed to validate the proposed algorithm. The PSNR (peak signal-to-noise ratio) of the proposed module is 29.420, and the SSIM (structural similarity) reaches 0.855. These results show that the proposed model works well for safety helmet detection in construction industries.

关键词： deep learning real-time detection safety helmet detection super-resolution reconstruction YOLOv5

来源：评论

学校读者我要写书评

暂无评论

Enhancing UAV tracking: a focus on discriminative representations using contrastive instances

引用

JOURNAL OF real-time image processing 2024年第3期21卷 78-78页

作者： Wang, Xucheng Zeng, Dan Li, Yongxin Zou, Mingliang Zhao, Qijun Li, Shuiwang Guilin Univ Technol Coll Comp Sci & Engn Guilin 541000 Peoples R China Southern Univ Sci & Technol Comp Sci & Engn Shenzhen 518000 Peoples R China Sichuan Univ Coll Comp Sci Chengdu 30332 Peoples R China

Addressing the core challenges of achieving both high efficiency and precision in UAV tracking is crucial due to limitations in computing resources, battery capacity, and maximum load capacity on UAVs. Discriminative correlation filter (DCF)-based trackers excel in efficiency on a single CPU but lag in precision. In contrast, many lightweight deep learning (DL)-based trackers based on model compression strike a better balance between efficiency and precision. However, higher compression rates can hinder performance by diminishing discriminative representations. Given these challenges, our paper aims to enhance feature representations' discriminative abilities through an innovative feature-learning approach. We specifically emphasize leveraging contrasting instances to achieve more distinct representations for effective UAV tracking. Our method eliminates the need for manual annotations and facilitates the creation and deployment of lightweight models. As far as our knowledge goes, we are the pioneers in exploring the possibilities of contrastive learning in UAV tracking applications. Through extensive experimentation across four UAV benchmarks, namely, UAVDT, DTB70, UAV123@10fps and VisDrone2018, We have shown that our DRCI (discriminative representation with contrastive instances) tracker outperforms current state-of-the-art UAV tracking methods, underscoring its potential to effectively tackle the persistent challenges in this field.

关键词： UAV tracking Contrastive instances Discriminative representation Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Activating high-frequency information nodes for super-resolution magnetic resonance imaging

引用

BIOMEDICAL SIGNAL processing AND CONTROL 2024年 93卷

作者： Li, Lei Liu, Yu Meng, Xiangshui Zhao, Yiming Wei, Shufeng Wang, Huixian Wang, Zheng Wei, Zhao Yang, Wenhui Chinese Acad Sci Inst Elect Engn 6 Beiertiao Beijing 100190 Peoples R China Univ Chinese Acad Sci Beijing Peoples R China Shandong Aoxin Med Technol Co LTD Weifang Peoples R China Shandong Univ Qilu Hosp Qingdao Cheeloo Coll Med Dept Radiol Qingdao Peoples R China

Objective: To recover the missing high-frequency components in low-resolution images by activating nodes in the high-frequency region of k-space. In addition, combining the information features in the k-space domain with the structural features in the image domain to generate high-quality super-resolution images. Methods: We propose a novel Dual-Domain cascaded Super -Resolution Network (DDSRNet), which shifts the starting point of the network to k-space. DDSRNet combines the ideas of GraphSAGE, Swin-Unet, and Hybrid Attention Transformer, to achieve complementary advantages of the k-space domain and the image domain. The k-space domain network activates zero-filled nodes in high-frequency regions and generates multi-order information features. The image domain network performs shallow, deep and gradient feature extraction to obtain the high-level representation of dual-domain hybrid features. In addition, to better utilize the features of both domains, we construct a domain interaction pool to facilitate cross-domain feature transfer and improve the efficiency of feature fusion. Results: Exhaustive experiments are conducted on both public dataset and real scanning dataset. Compared with state-of-the-art algorithms, DDSRNet has the best numerical evaluation results (the average PSNR and SSIM improvements are more than 0.1 dB and 0.0078) and visual perception. In addition, DDSRNet enables highquality 2x SR MRI with a reduced number of excitations, leading to increased low-field imaging speed. Conclusion: DDSRNet shows excellent performance in highand low-field SR tasks and has the potential to be a powerful tool for clinical applications. Significance: The proposed method is of great practical significance for achieving fast and high-quality magnetic resonance imaging.

关键词： MRI Super-resolution K-space Dual-domain deep learning

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：