A Moving Target Indication Algorithm Adaptable to Illuminance Variation Using the Histogram Matching
This paper addresses an illuminance compensation algorithm which can be used for small Unmanned Aerial Vehicles equipped with image sensors.A video sequence acquired in outdoor environments usually contains a time-varying illuminance.This illuminance variation between two image frames is an obstacle to change-detection algorithms such as Moving Target Indication(MTI)algorithm based on image sensors.We utilize the Histogram Matching method on the consecutive image frames in order to treat the illuminance variation problem.Illuminance models are generated from original images through the Gaussian filter,and their histograms are rearranged to match up with each other so that the algorithm minimizes the effect of the illuminance variation.To demonstrate the performance,the algorithm is applied to video sequences with a rapid illuminance change.
Histogram Matching Illuminance Compensation Moving Target Indication Computer Vision Object detection
Seung-Gyu Yun Sangchul Lee Sangho Ko
School of Aerospace and Mechamical Engineering,Korea Aerospace University,Goyang,Gyeonggi-do,412-791,Korea
国际会议
2014 Asia-Pacific International Symposium on Aerospace Technology(2014亚太航空航天技术学术会议)
上海
英文
1-5
2014-09-24（万方平台首次上网日期，不代表论文的发表时间）