SIA OpenIR  > 光电信息技术研究室
Alternative TitleBenchmark image quality evaluation method for heterogeneous image matching system
向伟; 惠斌; 花海洋; 王喆鑫; 王学娟; 刘云鹏
Rights Holder中国科学院沈阳自动化研究所
Patent Agent沈阳科苑专利商标代理有限公司 21002
Other AbstractThe method involves analyzing the influence of image features on heterogeneous image matching system and determining feature elements of benchmark image that affect matching performance of the matching system. The imaging geometry projection is used to design benchmark image space changing model to accurately calculate deformation of benchmark image characteristic under geometrical condition. The benchmark image multilevel code is designed and the performance of the image matching system is counted by image sample set to establish calibration library for evaluating quality of benchmark image. The similarity measure between the benchmark images is calculated by using mutual information similarity measure method to search for most similar benchmark image of benchmark image quality calibration library, obtain recognition performance calibration value of corresponding matching system and finish the quality evaluation of the benchmark image.
PCT Attributes
Application Date2016-12-20
Date Available2020-04-03
Application NumberCN201611181807.5
Open (Notice) NumberCN108205645A
Contribution Rank1
Document Type专利
Recommended Citation
GB/T 7714
向伟,惠斌,花海洋,等. 一种异源图像匹配系统的基准图质量评价方法[P]. 2018-06-26.
Files in This Item:
File Name/Size DocType Version Access License
CN201611181807.5.pdf(801KB)专利 开放获取CC BY-NC-SAView Application Full Text
CN201611181807.5授权.p(798KB)专利 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[向伟]'s Articles
[惠斌]'s Articles
[花海洋]'s Articles
Baidu academic
Similar articles in Baidu academic
[向伟]'s Articles
[惠斌]'s Articles
[花海洋]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[向伟]'s Articles
[惠斌]'s Articles
[花海洋]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: CN201611181807.5.pdf
Format: Adobe PDF
File name: CN201611181807.5授权.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.

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