检索结果-内蒙古大学图书馆

Blood cell counting based on U-Net++ and YOLOv5

optoelectronics Letters 2023年第6期19卷 370-376页

作者： BAI Hua WANG Xuechun GUAN Yingjian GAO Qiang HAN Zhibo Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems School of Electronic and Information EngineeringTiangong UniversityTianjin 300387China Shenzhen Weian Co-create Biotechnology Co. Ltd.Shenzhen 518000China Tianjin Key Laboratory of Engineering Technologies for Cell Phamaceutical Tianjin 300457China National Engineering Research Center of Cell Products/AmCellGene Co. Ltd.Tianjin 300457China

Clinical information about a variety of disorders is available through blood cell counting,which is usually done by manual ***,manual methods are complex,time-consuming and susceptible to the subjective experience of *** many efforts have been made to develop automated blood cell counting algorithms,the complexity of blood cell distribution and the highly overlapping nature of some red blood cells(RBCs) remain significant challenges that limit the improvement of analytical ***,we proposed an end-to-end method for blood cell counting based on deep ***,U-Net++ was used to segment the whole blood cell image into several regions of interest(ROI),and each ROI contains only one single cell or multiple overlapping ***,YOLOv5 was used to detect blood cells in each ***,we proposed several strategies,including fine classification of RBCs,adaptive adjustment for non-maximal suppression(NMS) threshold and blood cell morphology constraints to improve the accuracy of ***,the detection outcomes for each ROI were combined and *** results show that our method can effectively address the issue of high overlap and precisely segment and detect blood cells,with a 98.18% accuracy rate for blood cell counting.

关键词： Blood consuming overlap

来源：评论

学校读者我要写书评

暂无评论

CAD Cut-piece Retrieval Method Based on Representation Learning 2021

CAD Cut-piece Retrieval Method Based on Representation Learn...

引用

7th International Conference on Computing and Artificial Intelligence, ICCAI 2021

作者： Geng, Lei Yang, Yulong Xiao, Zhitao Yin, Changshun School of Life Sciences Tianjin Key Laboratory of Optoelectronic Detection Technology and System TianGong University Tianjin300387 China School of Electronics and Information Engineering Tianjin Key Laboratory of Optoelectronic Detection Technology and System Tiangong University Tianjin300387 China

ISBN: (纸本)9781450389501

The accuracy of retrieving the CAD file corresponding to the cut-piece from the CAD library determines the efficiency of the cut-piece automatic detection system and the digital management of the cut-piece inventory. However, the materials of the cut-piece are complex and diverse (including cowhide, artificial leather, fabric, etc.), contain multi-layered shapes and structures, and are prone to deformation. The traditional registration algorithms cannot satisfy the accurate registration between CAD and cut-piece. In order to solve the problem of high-frequency translation and complex shape of the cut-piece, we propose an image retrieval algorithm based on representation learning. This paper takes FaceNet as the basic framework. First, the loss function combining Softmax loss and center loss is used to maintain the discriminant ability between classes and optimize the cohesion of the ternary loss function. Second, the BlurPool module is used to enhance the anti-aliasing of the downsampling process and enhance the translation invariance of the network while maintaining network performance. Finally, the multi-feature deep fusion (FRE) module designed in this paper is used to solve the problem of reduced model expression ability caused by different sizes of cut-piece. A series of experiments on our dataset show that the proposed representation learning network is more accurate in the SVM classifier, and it only takes 30ms to predict a single cut-piece. © 2021 ACM.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Novel sparse planar array synthesis model for microwave power transmission systems with high efficiency and low cost

引用

Progress In Electromagnetics Research C 2021年 115卷 245-259页

作者： Li, Jianxiong Chang, Shengjia School of Electrical and Electronics Engineering Tiangong University Tianjin China Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems Tianjin China

—A novel algorithm is developed to realize the optimal synthesis of a sparse nonuniform-amplitude nonuniform-distribution planar array (SNANDPA) in microwave power transmission (MPT) systems. The dual compression factor particle swarm optimization (DCFPSO) algorithm and the subarray partition technique are adopted to realize the optimal synthesis of SNANDPA. The DCFPSO algorithm first optimizes beam collection efficiency (BCE) and side-lobe level outside the receiving region (CSL) of SNANDPA which ensure efficient energy transmission of an MPT system and suppress the influence of electromagnetic wave radiated by antenna array on the environment. The subarray partition technique then simplifies the feed network to minimize the system cost. SNANDPA parameters including transmitting aperture, receiving aperture, BCE, CSL, power pattern, element weight, and element distribution, can be obtained efficiently via the proposed method. Representative numerical cases under the different numbers of subarray and elements conditions are analyzed and compared with those of other two traditional MPT array models. Experimental results show that, when the transmitting aperture is 4.5λ × 4.5λ and the square receiving region u0 = v0 = 0.2, BCE and CSL are 94.96% and −17.09 dB, respectively, and only 64 elements and 8 amplifiers are required. We conclude that the proposed model can be used to create an efficient and low-cost MPT system. © 2021, Electromagnetics Academy. All rights reserved.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

Research on Instrument Image Generation Method Based on Conditional Generative Adversarial Network 2021

Research on Instrument Image Generation Method Based on Cond...

引用

7th International Conference on Computing and Artificial Intelligence, ICCAI 2021

作者： Geng, Lei Zhang, Yanyan Xiao, Zhitao Yan, Fengfeng School of Life Sciences TianGong University Tianjin Key Laboratory of Optoelectronic Detection Technology and System Tianjin300387 China School of Electronics and Information Engineering Tiangong University Tianjin Key Laboratory of Optoelectronic Detection Technology and System Tianjin300387 China

ISBN: (纸本)9781450389501

In recent years, generative adversarial networks have achieved great success in image *** have amazing ability to generate clear and realistic images. However, as we all know, GAN networks are difficult to adapt to different data sets, for example, in the process of image coding, the images of instrument are complex, instrument characters vary in size, and characters of different sizes need different receptive fields, so new requirements are put forward for the GAN networks. In this paper, we propose UYF-Net, an instrument image generator network, to solve this *** network uses Y-Fusion module many times, and fuses high-level and low-level semantic features according to the importance of each channel. Experiments show that the network achieves better performance in Mish activation function, mixed loss function of L1 with MS-SSIM and LayerNormal normalization. © 2021 ACM.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Phase Extraction of Electronic Speckle Interference Fringe Image based on Convolutional Neural Network 2021

Phase Extraction of Electronic Speckle Interference Fringe I...

引用

7th International Conference on Computing and Artificial Intelligence, ICCAI 2021

作者： Zhang, Fang Zhao, Rui Wang, Wen Liu, Yang School of Life Sciences Tianjin Key Laboratory of Optoelectronic Detection Technology and System TianGong University Tianjin300387 China School of Electronics and Information Engineering Tianjin Key Laboratory of Optoelectronic Detection Technology and System Tiangong University Tianjin300387 China

ISBN: (纸本)9781450389501

Electronic speckle pattern interferometry (ESPI) is a kind of full-field, non-contact nondestructive measurement technology, which is suitable for deformation measurement and nondestructive testing of optical rough surface. Fringe phase is an important information of interference fringe image, and the accuracy of phase estimation plays an important role in the extraction of fringe information. However, because of the discontinuity of fringe image and the influence of noise, phase extraction has always been a challenging problem. In this paper, we propose a fringe image phase extraction technique based on convolutional neural network, and experimental results show that this method has a good effect on phase estimation of the interference fringe. © 2021 ACM.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

SE-MIDNet Based on Deep Learning for Diabetic Retinopathy Classification 2021

SE-MIDNet Based on Deep Learning for Diabetic Retinopathy Cl...

引用

7th International Conference on Computing and Artificial Intelligence, ICCAI 2021

作者： Xiao, Zhitao Zhang, Yaxin Wu, Jun Zhang, Xinxin School of Life Sciences Tianjin Key Laboratory of Optoelectronic Detection Technology and System TianGong University Tianjin300387 China School of Electronics and Information Engineering Tianjin Key Laboratory of Optoelectronic Detection Technology and System TianGong University Tianjin300387 China

ISBN: (纸本)9781450389501

Diabetic Retinopathy (DR) is one of the most serious complications of diabetes. At present, DR detection mainly relies on detailed analysis by ophthalmologists. However, manual diagnosis is time-consuming and low efficiency. Aiming at the task of DR automatic classification, this paper proposes a classification method of DR based on deep learning. In view of the different sizes of the lesion area, firstly, an improved Inception module is proposed, which enables the network to efficiently extract multi-scale features of DR images. Then, the dense connection method is used to splice the output feature maps of the improved Inception module and send them to the subsequent layers to realize the multi-scale feature reuse of DR images and enhance the feature representation of small targets. Finally, the Squeeze-and-Excitation (SE) module is used to obtain the global information of the feature map on each channel, and the dynamic nonlinear modeling of each channel is carried out to improve the generalization ability of the network. The experimental results show that the network structure designed in this paper has good generalization ability, and the accuracy of DR automatic classification reaches 88.24%, the sensitivity reaches 99.43%, and the specificity reaches 97.6%, which can meet the needs of hospitals for DR classification. © 2021 ACM.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Research on an automatic light field refocusing method for a camera array

引用

Optics Express 2025年第10期33卷 20964-20980页

作者： Pan, Yihua Zhao, Yushan Li, Yuyang Liu, Bin Chen, Ping Shanxi Key Laboratory of Intelligent Detection Technology & Equipment School of Information and Communication Engineering North University of China Shanxi Taiyuan030051 China Key Laboratory of Photoelectronic Imaging Technology System of Ministry of Education of China School of Optics and Photonics Beijing Institute of Technology Beijing100081 China State Key Laboratory of Extreme Environment Optoelectronic Dynamic Testing Technology and Instruments North University of China Shanxi Taiyuan030051 China

Light field information includes the position and angle information of light in space. A camera array is one of the methods for obtaining light field information, which combines and arranges multiple cameras in space to achieve multi-angle imaging of the same target. Based on the light field information, a computational imaging method can be used to achieve refocusing technology for a camera array (also known as optical synthetic aperture imaging technology). In this technology, by setting the focusing coefficient, the light field information can be focused on different depths. This mode is called taking photos first and focusing later. However, in order to focus on a specific object surface, traditional refocusing methods require users to manually adjust the focusing coefficient multiple times, which is time-consuming and user-unfriendly. Therefore, in this paper, we propose an automatic focusing method based on the optimal focusing coefficient map (OFCM) to address the manual refocusing problem. This method transforms the refocusing problem into obtaining the optimal focus coefficient map. In the process of solving OFCM, the threshold filtering method and the neighborhood filling method are used to optimize the results and improve the robustness of the proposed method. The experimental results show that our proposed method has higher imaging quality and faster running efficiency compared to automatic refocusing methods based on disparity maps and gradient functions. © 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.

关键词： Computational Imaging

来源：评论

学校读者我要写书评

暂无评论

Automatic detection of Internal Defects of Composite Materials based on ESPI 2021

Automatic Detection of Internal Defects of Composite Materia...

引用

7th International Conference on Computing and Artificial Intelligence, ICCAI 2021

作者： Zhang, Fang Zhang, Qian Xiao, ZhiTao Geng, Lei School of Life Sciences Tianjin Key Laboratory of Optoelectronic Detection Technology and System TianGong University Tianjin300387 China School of Electronics and Information Engineering Tianjin Key Laboratory of Optoelectronic Detection Technology and System Tiangong University Tianjin300387 China

ISBN: (纸本)9781450389501

Composite materials will inevitably have defects and suffer damage during the manufacturing and using. In order to achieve accurate and efficient non-destructive testing of composite materials, an automatic detection system for internal defects of composite materials based on electronic speckle pattern interferometry (ESPI) is designed. The system uses the collected speckle stripes as the defect carrier, and realizes the detection of defects in the composite material by identifying the speckle stripes image. First of all, in order to expand the field of view of the speckle collection light path, we complete the construction of the collection system hardware light path combining the 4f system, and realizes a miniaturized design. Then, the speckle image acquisition control module was designed and combined with the miniaturized instrument to realize the visualization of speckle interference fringes. Finally, in order to realize automatic defect detection, a target detection model is introduced to identify and locate speckle fringes. Experiments have confirmed that the automatic detection of internal defects of composite materials was well completed. The designed miniaturized speckle acquisition system is suitable for defect detection of composite materials. The defect detection network model shows good fringe detection performance, and its mAP reaches 97.91%. © 2021 ACM.

关键词： Speckle

来源：评论

学校读者我要写书评

暂无评论

Label-Free Evaluation of Treatment Efficacy in Laser Therapy by Terahertz Imaging

Label-Free Evaluation of Treatment Efficacy in Laser Therapy...

引用

Cross Strait Quad-Regional Radio Science and Wireless technology Conference (CSRWTC)

作者： Cong Xu Guanlong Wang Jia Shi Tianjin Key Laboratory of Optoelectronic Detection Technology and System School of Electronic and Information Engineering Tiangong University Tianjin China

Low-level laser therapy (LLLT) is a promising non-invasive treatment method in biomedical science. Rapid and high-sensitivity imaging evaluation of treatment efficacy in LLLT always remains changeling. To address this issue, here the confocal terahertz imaging is used to label-free evaluation of treatment efficacy in LLLT on ALI. The results show that terahertz imaging can effectively evaluate the therapeutic effect of LLLT on ALI with the advantages of high sensitivity.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Resin-Based Dielectric Terahertz Metalens and Applications

Resin-Based Dielectric Terahertz Metalens and Applications

引用

Cross Strait Quad-Regional Radio Science and Wireless technology Conference (CSRWTC)

作者： Jia Shi Guanlong Wang Cong Xu Tianjin Key Laboratory of Optoelectronic Detection Technology and System School of Electronic and Information Engineering Tiangong University Tianjin China

All-dielectric metalenses have become a research hotspot because of their high-efficiency and customizable features. Here, a resin-based dielectric metalens is designed by two-dimensional circumferential metagrating arrays for terahertz imaging beyond classical physical diffraction limit. The metalens is fabricated with UV -sensitive resin by a photo curing 3D printer. With the proposed scheme, the terahertz beam can be focused as a filamentous focal spot in sub-wavelength scale with a long depth of focus. As its demonstration in terahertz imaging applications, the printed transmission lines on the PCB board can be clearly imaged. The proposed metalens shows great potential in high-resolution 6G terahertz spatial sensing and imaging.

关键词：

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：