IOE OpenIR  > 光电技术研究所博硕士论文
空间目标三维姿态提取方法研究
Alternative TitleResearch on Algorithms of Extracting the 3D Pose Information of Space Targets
杨赟
Subtype硕士
Thesis Advisor周进
2009-05-25
Degree Grantor中国科学院光电技术研究所
Place of Conferral光电技术研究所
Degree Discipline信号与信息处理
Keyword图像增强 边缘检测 直线增强 姿态测量 Snake模型 Gac
Abstract基于数字图像处理方法,本文设计了目标姿态判读算法。利用该算法对光学设备记录的图像进行处理,能够自动、稳定、高精度地提取出目标在图像中的姿态信息。再利用两站或多站的姿态测量结果进行交会计算,可以获取空间目标的三维姿态参数。 本文根据空间目标的特征,设计了基于边缘检测和Hough变换直线提取的目标姿态提取算法。 首先根据实际图像特点和侧抑制效应的原理,设计了一种适用于光照不均匀图像的图像增强算法。 然后对Canny边缘检测算法进行改进,并结合直线边缘增强算法,使之更适合于低对比度图像中直线边缘的检测。 在获取目标边缘后,利用Hough变换提取直线边缘。通过分析目标直线边缘特征,基于树分类法的原理,设计了满足实际工程要求的目标识别算法。基于Snake主动轮廓模型的原理,设计直线Snake模型,对识别出的目标直线边缘进行修正,然后计算获取目标的高精度姿态参数。 为了进一步提高目标姿态提取算法的稳定性、自动化程度和精度。设计了一种新的能量泛函对几何主动轮廓模型(GAC)进行改进,并基于其设计新的目标标姿态提取算法,实验结果表明算法的性能十分优异。
Other AbstractThis paper presents algorithms of extracting the target’s pose information based on digital image processing technology, which can measure the target’s pose information in images from optical equipments automatically and stably in high precision. With data from two or more measuring stations, the target’s 3D pose information can be obtained by intersecting calculation. Based on the characteristics of the target, a pose extracting algorithm based on edge detection and Hough line extracting method is presented. First,based on the characteristics of real images and the lateral inhibition mechanism, an image enhancing method which is suitable for non-uniform illuminated images is presented. Then an improved Canny edge detector and a novel straight line enhancing method is presented, which are suitable for detecting straight line edges in low contrast images. After getting the edges, the line edges are extracted by Hough line extracting method. Through analyzing the characteristics of the targets, a target recognizing method based on the theory of Tree-Classifying Method is presented, which satisfies the requirement of practical projects. Then, the Straight Line Snake model based on the theory of Snake active contour model is presented to modify the target’s straight-line edges. Finally, the target’s pose information is calculated accurately. In order to extract the target’s pose information more stably and automatically and in higher precision, a new pose extracting algorithm is presented, which is based on the Geometric Active Contour model (GAC) improved by a novel energy function. The experimental results show that the new algorithm has excellent performance.
Pages66
Language中文
Document Type学位论文
Identifierhttp://ir.ioe.ac.cn/handle/181551/378
Collection光电技术研究所博硕士论文
Recommended Citation
GB/T 7714
杨赟. 空间目标三维姿态提取方法研究[D]. 光电技术研究所. 中国科学院光电技术研究所,2009.
Files in This Item:
File Name/Size DocType Version Access License
10001_20062801512705(2403KB) 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[杨赟]'s Articles
Baidu academic
Similar articles in Baidu academic
[杨赟]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杨赟]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.