期刊文献+
共找到1,487篇文章
< 1 2 75 >
每页显示 20 50 100
Automated Extraction and Analysis of CBC Test from Scanned Images
1
作者 Iman S. Alansari 《Journal of Software Engineering and Applications》 2024年第2期129-141,共13页
Health care is an important part of human life and is a right for everyone. One of the most basic human rights is to receive health care whenever they need it. However, this is simply not an option for everyone due to... Health care is an important part of human life and is a right for everyone. One of the most basic human rights is to receive health care whenever they need it. However, this is simply not an option for everyone due to the social conditions in which some communities live and not everyone has access to it. This paper aims to serve as a reference point and guide for users who are interested in monitoring their health, particularly their blood analysis to be aware of their health condition in an easy way. This study introduces an algorithmic approach for extracting and analyzing Complete Blood Count (CBC) parameters from scanned images. The algorithm employs Optical Character Recognition (OCR) technology to process images containing tabular data, specifically targeting CBC parameter tables. Upon image processing, the algorithm extracts data and identifies CBC parameters and their corresponding values. It evaluates the status (High, Low, or Normal) of each parameter and subsequently presents evaluations, and any potential diagnoses. The primary objective is to automate the extraction and evaluation of CBC parameters, aiding healthcare professionals in swiftly assessing blood analysis results. The algorithmic framework aims to streamline the interpretation of CBC tests, potentially improving efficiency and accuracy in clinical diagnostics. 展开更多
关键词 image processing Optical Character Recognition Tesseract OCR Health Care Application
下载PDF
Influences of Atmospheric Turbulence on Image Resolution of Airborne and Space-Borne Optical Remote Sensing System 被引量:2
2
作者 张晓芳 俞信 阎吉祥 《Journal of Beijing Institute of Technology》 EI CAS 2006年第4期457-461,共5页
A new way is proposed to evaluate the influence of atmospheric turbulence on image resolution of airborne and space-borne optical remote sensing system, which is called as arrival angle-method. Applying this method, s... A new way is proposed to evaluate the influence of atmospheric turbulence on image resolution of airborne and space-borne optical remote sensing system, which is called as arrival angle-method. Applying this method, some engineering examples are selected to analyze the turbulence influences on image resolution based on three different atmospheric turbulence models quantificationally, for the airborne remote sensing system, the resolution errors caused by the atmospheric turbulence are less than 1 cm, and for the space-borne remote sensing system, the errors are around 1 cm. The results are similar to that obtained by the previous Friedmethod. Compared with the Fried-method, the arrival angle-method is rather simple and can be easily used in engineering fields. 展开更多
关键词 atmospheric turbulence coherence length arrival angle-method airborne or space-borne optical remote sensing system image resolution
下载PDF
Radial Basis Function Neural Network Based Super- Resolution Restoration for an Undersampled Image 被引量:1
3
作者 苏秉华 金伟其 牛丽红 《Journal of Beijing Institute of Technology》 EI CAS 2004年第2期135-138,共4页
To achieve restoration of high frequency information for an undersampled and degraded low-resolution image, a nonlinear and real-time processing method-the radial basis function (RBF) neural network based super-resolu... To achieve restoration of high frequency information for an undersampled and degraded low-resolution image, a nonlinear and real-time processing method-the radial basis function (RBF) neural network based super-resolution method of restoration is proposed. The RBF network configuration and processing method is suitable for a high resolution restoration from an undersampled low-resolution image. The soft-competition learning scheme based on the k-means algorithm is used, and can achieve higher mapping approximation accuracy without increase in the network size. Experiments showed that the proposed algorithm can achieve a super-resolution restored image from an undersampled and degraded low-resolution image, and requires a shorter training time when compared with the multiplayer perception (MLP) network. 展开更多
关键词 SUPER-RESOLUTION image restoration image processing neural networks UNDERSAMPLING
下载PDF
Anisotropic Total Variation Regularization Based NAS-RIF Blind Restoration Method for OCT Image 被引量:2
4
作者 Xuesong Fu Jianlin Wang +3 位作者 Zhixiong Hu Yongqi Guo Kepeng Qiu Rutong Wang 《Journal of Beijing Institute of Technology》 EI CAS 2020年第2期146-157,共12页
Based on anisotropic total variation regularization(ATVR), a nonnegativity and support constraints recursive inverse filtering(NAS-RIF) blind restoration method is proposed to enhance the quality of optical coherence ... Based on anisotropic total variation regularization(ATVR), a nonnegativity and support constraints recursive inverse filtering(NAS-RIF) blind restoration method is proposed to enhance the quality of optical coherence tomography(OCT) image. First, ATVR is introduced into the cost function of NAS-RIF to improve the noise robustness and retain the details in the image.Since the split Bregman iterative is used to optimize the ATVR based cost function, the ATVR based NAS-RIF blind restoration method is then constructed. Furthermore, combined with the geometric nonlinear diffusion filter and the Poisson-distribution-based minimum error thresholding, the ATVR based NAS-RIF blind restoration method is used to realize the blind OCT image restoration. The experimental results demonstrate that the ATVR based NAS-RIF blind restoration method can successfully retain the details in the OCT images. In addition, the signal-to-noise ratio of the blind restored OCT images can be improved, along with the noise robustness. 展开更多
关键词 optical coherence tomography(OCT)image blind image restoration cost function nonnegativity and support constraints recursive inverse filtering(NAS-RIF)
下载PDF
RESTORATION OF THE IMAGE DEGRADED BY LINEAR MOTION 被引量:1
5
作者 李允明 竺卫东 《Journal of China Textile University(English Edition)》 EI CAS 1990年第4期27-36,共10页
This paper introduces a new effective method to restore the uniform linear motion blurred im-age. The effect of the out-of-frame pixels on the blurring process and the estimate of these pixelsare analysed. The restora... This paper introduces a new effective method to restore the uniform linear motion blurred im-age. The effect of the out-of-frame pixels on the blurring process and the estimate of these pixelsare analysed. The restoration qualities of different deblurring methods are compared. Finally, theauthors come to a conclusion that it is impossible to determine the length of blurring movement infrequency domain. 展开更多
关键词 image processing image RESTITUTION methods blurred image image restoration
下载PDF
Improvement Detecting Method of Optical Axes Parallelism of Shipboard Photoelectrical Theodolite Based on Image Processing 被引量:3
6
作者 Huihui Zou 《Optics and Photonics Journal》 2017年第8期127-133,共7页
An improvement detecting method was proposed according to the disadvantages of testing method of optical axes parallelism of shipboard photoelectrical theodolite (short for theodolite) based on image processing. Point... An improvement detecting method was proposed according to the disadvantages of testing method of optical axes parallelism of shipboard photoelectrical theodolite (short for theodolite) based on image processing. Pointolite replaced 0.2'' collimator to reduce the errors of crosshair images processing and improve the quality of image. What’s more, the high quality images could help to optimize the image processing method and the testing accuracy. The errors between the trial results interpreted by software and the results tested in dock were less than 10'', which indicated the improve method had some actual application values. 展开更多
关键词 IMPROVEMENT Detecting Method SHIPBOARD Photoelectrical THEODOLITE OPTICAL Axes PARALLELISM image processing
下载PDF
Automated cone photoreceptor cell identication in confocal adaptive optics scanning laser ophthalmoscope images based on object detection 被引量:1
7
作者 Yiwei Chen Yi He +4 位作者 Jing Wang Wanyue Li Lina Xing Xin Zhang Guohua Shi 《Journal of Innovative Optical Health Sciences》 SCIE EI CAS 2022年第1期103-109,共7页
Cone photoreceptor cell identication is important for the early diagnosis of retinopathy.In this study,an object detection algorithm is used for cone cell identication in confocal adaptive optics scanning laser ophtha... Cone photoreceptor cell identication is important for the early diagnosis of retinopathy.In this study,an object detection algorithm is used for cone cell identication in confocal adaptive optics scanning laser ophthalmoscope(AOSLO)images.An effectiveness evaluation of identication using the proposed method reveals precision,recall,and F_(1)-score of 95.8%,96.5%,and 96.1%,respectively,considering manual identication as the ground truth.Various object detection and identication results from images with different cone photoreceptor cell distributions further demonstrate the performance of the proposed method.Overall,the proposed method can accurately identify cone photoreceptor cells on confocal adaptive optics scanning laser ophthalmoscope images,being comparable to manual identication. 展开更多
关键词 Biomedical image processing retinal imaging adaptive optics scanning laser ophthalmoscope object detection.
下载PDF
Application of improved BPNN in image restoration-learning coefficient
8
作者 Umar Farooq 沈庭芝 +3 位作者 Muhammad Imran 赵三元 Sadia Murawwat 王清云 《Journal of Beijing Institute of Technology》 EI CAS 2012年第4期543-546,共4页
A new method of artificial intelligence based on a new improved back propagation neural network (BPNN) algorithm is partially applied in the problem of image restoration. In order to over- come the inherited issues ... A new method of artificial intelligence based on a new improved back propagation neural network (BPNN) algorithm is partially applied in the problem of image restoration. In order to over- come the inherited issues in conventional back propagation algorithm i.e. slow convergence rate, longer training time, hard to achieve global minima etc. , different methods have been used including the introduction of dynamic learning rate and dynamic momentum coefficient etc. With the passage of time different techniques has been used to improve the dynamicity of these coefficients. The meth- od applied in this paper improves the effect of learning coefficient η by using a new way to modify the value dynamically during learning process. The experimental results show that this helps in im- proving the efficiency overall both in visual effect and quality analysis. 展开更多
关键词 image restoration image processing INTELLIGENT back propagation neural network(BPNN) dynamic learning coefficient
下载PDF
Single-Phase Velocity Determination Based in Video and Sub-Images Processing:An Optical Flow Method Implemented with Support of a Programmed MatLab Structured Script 被引量:1
9
作者 Andreas Nascimento Edson Da Costa Bortoni +2 位作者 José Luiz Goncalves Pedro Antunes Duarte Mauro Hugo Mathias 《Journal of Software Engineering and Applications》 2015年第6期290-294,共5页
Important in many different sectors of the industry, the determination of stream velocity has become more and more important due to measurements precision necessity, in order to determine the right production rates, d... Important in many different sectors of the industry, the determination of stream velocity has become more and more important due to measurements precision necessity, in order to determine the right production rates, determine the volumetric production of undesired fluid, establish automated controls based on these measurements avoiding over-flooding or over-production, guaranteeing accurate predictive maintenance, etc. Difficulties being faced have been the determination of the velocity of specific fluids embedded in some others, for example, determining the gas bubbles stream velocity flowing throughout liquid fluid phase. Although different and already applicable methods have been researched and already implemented within the industry, a non-intrusive automated way of providing those stream velocities has its importance, and may have a huge impact in projects budget. Knowing the importance of its determination, this developed script uses a methodology of breaking-down real-time videos media into frame images, analyzing by pixel correlations possible superposition matches for further gas bubbles stream velocity estimation. In raw sense, the script bases itself in functions and procedures already available in MatLab, which can be used for image processing and treatments, allowing the methodology to be implemented. Its accuracy after the running test was of around 97% (ninety-seven percent);the raw source code with comments had almost 3000 (three thousand) characters;and the hardware placed for running the code was an Intel Core Duo 2.13 [Ghz] and 2 [Gb] RAM memory capable workstation. Even showing good results, it could be stated that just the end point correlations were actually getting to the final solution. So that, making use of self-learning functions or neural network, one could surely enhance the capability of the application to be run in real-time without getting exhaust by iterative loops. 展开更多
关键词 Optical Flow Single-Phase Velocity Video and image processing Sensing MatLab Script
下载PDF
A GAUSSIAN MIXTURE MODEL-BASED REGULARIZATION METHOD IN ADAPTIVE IMAGE RESTORATION
10
作者 Liu Peng Zhang Yan Mao Zhigang 《Journal of Electronics(China)》 2007年第1期83-89,共7页
A GMM (Gaussian Mixture Model) based adaptive image restoration is proposed in this paper. The feature vectors of pixels are selected and extracted. Pixels are clustered into smooth,edge or detail texture region accor... A GMM (Gaussian Mixture Model) based adaptive image restoration is proposed in this paper. The feature vectors of pixels are selected and extracted. Pixels are clustered into smooth,edge or detail texture region according to variance-sum criteria function of the feature vectors. Then pa-rameters of GMM are calculated by using the statistical information of these feature vectors. GMM predicts the regularization parameter for each pixel adaptively. Hopfield Neural Network (Hopfield-NN) is used to optimize the objective function of image restoration,and network weight value matrix is updated by the output of GMM. Since GMM is used,the regularization parameters share properties of different kind of regions. In addition,the regularization parameters are different from pixel to pixel. GMM-based regularization method is consistent with human visual system,and it has strong gener-alization capability. Comparing with non-adaptive and some adaptive image restoration algorithms,experimental results show that the proposed algorithm obtains more preferable restored images. 展开更多
关键词 image processing Gaussian Mixture Model (GMM) Hopfield Neural Network (Hopfield-NN) REGULARIZATION Adaptive image restoration
下载PDF
DECONVOLUTION IN TRANSFORM-DOMAIN FOR IMAGE DATA RESTORATION
11
作者 SunXiaojun DingQun 《Journal of Electronics(China)》 2005年第3期312-314,共3页
A novel scheme for image data restoration is proposed in this letter. First, a window- function model is exploited to describe the data loss in images. It can change the restoration problem into deconvolution in trans... A novel scheme for image data restoration is proposed in this letter. First, a window- function model is exploited to describe the data loss in images. It can change the restoration problem into deconvolution in transform-domain. Then, an iterative algorithm is presented to solve the deconvolution. Because the window-function is available to describe arbitrary shape, our algorithm is suitable for restoring irregular segment of data loss, including square-block. Finally, several simulation tests are done and results prove that the algorithm is valid. 展开更多
关键词 image processing Data restoration DECONVOLUTION
下载PDF
New Optical Triangulation and Digital Image Processing in Measurement
12
作者 ZHOU Jian ZHAO Hong +2 位作者 CHEN Wenyi TIAN Feng TAN Yushan(Xi’an Jiaotong University,Xi’an 710049 CHN) 《Semiconductor Photonics and Technology》 CAS 1996年第2期103-107,共5页
In this paper,a new direct optical triangulation(DOT) for measuring theout-of-plane displacement is given.In order to state its principle,DOT is used to measure a micro-displacement of a rigid body,and at the same tim... In this paper,a new direct optical triangulation(DOT) for measuring theout-of-plane displacement is given.In order to state its principle,DOT is used to measure a micro-displacement of a rigid body,and at the same time,the method of digital image processing is also given. 展开更多
关键词 Optical Triangulation Displacement Measurement SENSOR Digital image processing
下载PDF
Image Restoration of Depth of Field Extension Imaging System Based on Genetic Algorithm
13
作者 He Yun Wu Yangang +1 位作者 Tian Jialin Xu Wen 《International Journal of Technology Management》 2013年第2期37-40,共4页
Genetic algorithm is a search algorithm based on genetic mechanism and natural selection. It has been widely applied to research fields including image processing field. The paper improves standard genetic algorithm a... Genetic algorithm is a search algorithm based on genetic mechanism and natural selection. It has been widely applied to research fields including image processing field. The paper improves standard genetic algorithm and improves the arithmetic speed of the algorithm, which achieves better image restoration effect. And the paper compares the image restoration quality of traditional algorithm, standard genetic algorithm and improved genetic algorithm to prove the feasibility of applying genetic algorithm to image restoration. 展开更多
关键词 image restoration genetic algorithm image processing image degradation
下载PDF
Optical Image Encryption Based on Mixed Chaotic Maps and Single-Shot Digital Holography 被引量:3
14
作者 Yonggang Su Chen Tang +3 位作者 Xia Chen Biyuan Li Wenjun Xu Zhenkun Lei 《Transactions of Tianjin University》 EI CAS 2017年第2期184-191,共8页
Random phase masks play a key role in optical image encryption schemes based on double random phase technique. In this paper, a mixed chaotic method is proposed, which can efficiently solve some weaknesses that one-di... Random phase masks play a key role in optical image encryption schemes based on double random phase technique. In this paper, a mixed chaotic method is proposed, which can efficiently solve some weaknesses that one-dimensional (1-D) single chaotic maps encounter to generate random phase masks. Based on the chaotic random phase masks, optical image encryption and decryption are realized with a single-shot digital holographic technique. In the proposed encryption scheme, the initial value and parameters of mixed chaotic maps serve as secret keys, which is convenient for the key management and transmission. Moreover, it also possesses high resistance against statistical attack, brute-force attack, noise attack and shear attack. Simulation results and security analysis verify the validity and security of the proposed encryption scheme. © 2017, Tianjin University and Springer-Verlag Berlin Heidelberg. 展开更多
关键词 Chaotic systems Geometrical optics HOLOGRAPHY image processing Lyapunov methods Optical data processing
下载PDF
Artifacts Reduction Using Multi-Scale Feature Attention Network in Compressed Medical Images 被引量:1
15
作者 Seonjae Kim Dongsan Jun 《Computers, Materials & Continua》 SCIE EI 2022年第2期3267-3279,共13页
Medical image compression is one of the essential technologies to facilitate real-time medical data transmission in remote healthcare applications.In general,image compression can introduce undesired coding artifacts,... Medical image compression is one of the essential technologies to facilitate real-time medical data transmission in remote healthcare applications.In general,image compression can introduce undesired coding artifacts,such as blocking artifacts and ringing effects.In this paper,we proposed a Multi-Scale Feature Attention Network(MSFAN)with two essential parts,which are multi-scale feature extraction layers and feature attention layers to efficiently remove coding artifacts of compressed medical images.Multiscale feature extraction layers have four Feature Extraction(FE)blocks.Each FE block consists of five convolution layers and one CA block for weighted skip connection.In order to optimize the proposed network architectures,a variety of verification tests were conducted using validation dataset.We used Computer Vision Center-Clinic Database(CVC-ClinicDB)consisting of 612 colonoscopy medical images to evaluate the enhancement of image restoration.The proposedMSFAN can achieve improved PSNR gains as high as 0.25 and 0.24 dB on average compared to DnCNNand DCSC,respectively. 展开更多
关键词 Medical image processing convolutional neural network deep learning TELEMEDICINE artifact reduction image restoration
下载PDF
A single image dehazing method based on decomposition strategy 被引量:1
16
作者 QIN Chaoxuan GU Xiaohui 《Journal of Systems Engineering and Electronics》 SCIE EI CSCD 2022年第2期279-293,共15页
Outdoor haze has adverse impact on outdoor image quality,including contrast loss and poor visibility.In this paper,a novel dehazing algorithm based on the decomposition strategy is proposed.It combines the advantages ... Outdoor haze has adverse impact on outdoor image quality,including contrast loss and poor visibility.In this paper,a novel dehazing algorithm based on the decomposition strategy is proposed.It combines the advantages of the two-dimensional variational mode decomposition(2DVMD)algorithm and dark channel prior.The original hazy image is adaptively decom-posed into low-frequency and high-frequency images according to the image frequency band by using the 2DVMD algorithm.The low-frequency image is dehazed by using the improved dark channel prior,and then fused with the high-frequency image.Furthermore,we optimize the atmospheric light and transmit-tance estimation method to obtain a defogging effect with richer details and stronger contrast.The proposed algorithm is com-pared with the existing advanced algorithms.Experiment results show that the proposed algorithm has better performance in comparison with the state-of-the-art algorithms. 展开更多
关键词 single image dehazing decomposition strategy image processing global atmospheric light
下载PDF
UFC-Net with Fully-Connected Layers and Hadamard Identity Skip Connection for Image Inpainting
17
作者 Chung-Il Kim Jehyeok Rew +1 位作者 Yongjang Cho Eenjun Hwang 《Computers, Materials & Continua》 SCIE EI 2021年第9期3447-3463,共17页
Image inpainting is an interesting technique in computer vision and artificial intelligence for plausibly filling in blank areas of an image by referring to their surrounding areas.Although its performance has been im... Image inpainting is an interesting technique in computer vision and artificial intelligence for plausibly filling in blank areas of an image by referring to their surrounding areas.Although its performance has been improved significantly using diverse convolutional neural network(CNN)-based models,these models have difficulty filling in some erased areas due to the kernel size of the CNN.If the kernel size is too narrow for the blank area,the models cannot consider the entire surrounding area,only partial areas or none at all.This issue leads to typical problems of inpainting,such as pixel reconstruction failure and unintended filling.To alleviate this,in this paper,we propose a novel inpainting model called UFC-net that reinforces two components in U-net.The first component is the latent networks in the middle of U-net to consider the entire surrounding area.The second component is the Hadamard identity skip connection to improve the attention of the inpainting model on the blank areas and reduce computational cost.We performed extensive comparisons with other inpainting models using the Places2 dataset to evaluate the effectiveness of the proposed scheme.We report some of the results. 展开更多
关键词 image processing computer vision image inpainting image restoration generative adversarial nets
下载PDF
IMAGE-BASED IN VIVO QUANTITATIVE ASSESSMENT OF HUMAN AIRWAY OPENING AND CONTR ACTILITY BY FIBER OPTICAL NA SOPH A RYNGOSCOPY IN HEALTHY AND ASTHMATIC SUBJECTS
18
作者 LINHONG DENG 《Journal of Innovative Optical Health Sciences》 SCIE EI CAS 2013年第2期51-62,共12页
Assessment of human airway humen opening is important in diagnosing and understanding the mechanisms of airway dysfunctions such as the excessive airway narrowing in asthma and chronic obstructive pulmonary disease(CO... Assessment of human airway humen opening is important in diagnosing and understanding the mechanisms of airway dysfunctions such as the excessive airway narrowing in asthma and chronic obstructive pulmonary disease(COPD).Although there are indirect methods to evaluate the airway calibre,direct in vivo measurement of the airway calibre has not been commonly available.With recent advent of the flexible fiber optical nasopharyngoscope with video recording it has become possible to directly visualize the passages of upper and lower airways.However,quan-titative analysis of the recorded video images has been technically challenging.Here,we describe an automatic image processing and analysis method that allows for batch analysis of the images recorded during the endoscopic procedure,thus facilitates image-based quantification of the airway opening.Video images of the airway lumen of volunteer subject were acquired using a fiber optical nasopharyngoscope,and subsequently processed using Gaussian smoothing filter,threshold segment ation,differentiation,and Canny image edge detection,respectively.Thus the area of the open airway lumen was identified and computed using.a predetermined converter of the image scale to true dimension of the imaged object.With this method we measured the opening/narrowing of the glottis during tidal breathing with or without making“Hee"sound or cough.We also used this met hod to measure the opening/narrowing of the primary bronchus of either healthy or asthmatic subjects in response to hist amine and/or albuterol treatment,which also provided an indicator of the airway contractility.Our results demonstrate that the image-based method accurately quantifed the area change waveform of either the glottis or the bronchus as observed by using the optical nasopharygoscope.Importantly,the opening/nar-rowing of the airway lumen generally correlated with the airAow and resistance of the airways,and could differentiate the level of airway contr actility between the healthy and asthmatic subjects.Thus,this quant itative assessment of airway opening may provide a useful tool to ssist clinical diagnosis of airway dysfunctions and understanding the mechanisms of associated pathophysiologies. 展开更多
关键词 Optical nasophary ngoscopy image processing glottal aperture bronchus opening airway contractility asthma
下载PDF
Optimizing photoacoustic image reconstruction using cross-platform parallel computation
19
作者 Tri Vu Yuehang Wang Jun Xia 《Visual Computing for Industry,Biomedicine,and Art》 2018年第1期12-17,共6页
Three-dimensional(3D)image reconstruction involves the computations of an extensive amount of data that leads to tremendous processing time.Therefore,optimization is crucially needed to improve the performance and eff... Three-dimensional(3D)image reconstruction involves the computations of an extensive amount of data that leads to tremendous processing time.Therefore,optimization is crucially needed to improve the performance and efficiency.With the widespread use of graphics processing units(GPU),parallel computing is transforming this arduous reconstruction process for numerous imaging modalities,and photoacoustic computed tomography(PACT)is not an exception.Existing works have investigated GPU-based optimization on photoacoustic microscopy(PAM)and PACT reconstruction using compute unified device architecture(CUDA)on either C++or MATLAB only.However,our study is the first that uses cross-platform GPU computation.It maintains the simplicity of MATLAB,while improves the speed through CUDA/C++−based MATLAB converted functions called MEXCUDA.Compared to a purely MATLAB with GPU approach,our cross-platform method improves the speed five times.Because MATLAB is widely used in PAM and PACT,this study will open up new avenues for photoacoustic image reconstruction and relevant real-time imaging applications. 展开更多
关键词 Photoacoustic computed tomography Graphics processing units Parallel computation Focal-line backprojection algorithm MATLAB Optical imaging
下载PDF
基于改进Deformable-DETR的水下图像目标检测方法 被引量:2
20
作者 崔颖 韩佳成 +1 位作者 高山 陈立伟 《应用科技》 CAS 2024年第1期30-36,91,共8页
针对由于水下复杂环境造成的目标检测效果较差、检测精度较低的问题,基于Deformable-DETR算法提出一种改进的水下目标检测算法Deformable-DETR-DA。使用空间注意力模块结合标准Transformer块设计了一个用于增加模型深度的深度特征金字塔... 针对由于水下复杂环境造成的目标检测效果较差、检测精度较低的问题,基于Deformable-DETR算法提出一种改进的水下目标检测算法Deformable-DETR-DA。使用空间注意力模块结合标准Transformer块设计了一个用于增加模型深度的深度特征金字塔(deep feature pyramid networks,DFPN)模块,将其嵌入到模型中提高模型对深层纹理信息的提取能力。使用注意力引导的方式对原模型中编码器部分进行改进,加强了对特征信息的聚合能力,提高了模型在复杂环境下的检测能力。针对URPC数据集,模型各交并比尺度的平均准确度(average precision,AP)为39.5%,相比原模型提升1%,与一些DETR(detection transformer)类的模型相比,不同目标尺度的平均准确度均有1%~4%左右的提高,表明改进的模型能够很好解决复杂环境的水下目标检测的问题。本文提出的模型可作为其他水下目标检测模型设计的参考。 展开更多
关键词 水下光学图像 Deformable-DETR 目标检测 TRANSFORMER 注意力机制 深度学习 图像处理 残差网络
下载PDF
上一页 1 2 75 下一页 到第
使用帮助 返回顶部