Color Balance Method for Digital Orthographic Maps based on Neural Network
Color and radiation must be balanced prior to image mosaics. A color balance method for digital orthographic maps based on neural network is introduced in this paper. According to the proposed method two well-registered and contiguous DOM are handled as two kinds of one as standard image and the other target. It treated the overlapped region as inputs and outputs, and then made use of neural network to model the relationship between standard image and target image. Through the relationship obtained the target image was balanced entirely. Experiments prove that two images balanced in this way can be homogenous in color and uniform in tone.
Cao Zhaohui Shi Ruizhi
Zhengzhou Institute of Surveying and Mapping Zhengzhou, Henan Province, P.R. China
国际会议
The 31st International Congress on Imaging Science(第31届国际影像科学大会 ICIS2010)
北京
英文
69-71
2010-05-12（万方平台首次上网日期，不代表论文的发表时间）