Transfer Textures and Colors to Low Light Level Images

摘要：

In this paper, we propose a method for texture and color enhancement of low light level (LLL) night vision images. We design a texture database in accordance with the features of LLL images based on mean shift smoothing and luminance transformation. Each pixel in an LLL image is classified using this texture database. After the classification, we use a novel texture and color overprint method to assign consistent textures and colors to the LLL image. That is, we replace the pixels of each class in the LLL image by the pixels in another high-resolution color image. The proposed method is simple and it does not require any color space transformation. It overcomes the achromatic, low resolution, low contrast and lack of fine textures problem of the LLL night vision images and improves the situational awareness of night vision images.

关键词： texture and color transfer image overprint color night vision LLL image

作者: Shaoyuan Sun Haitao Zhao Xiaojing Gu

作者单位: Automation Department, Donghua University,Shanghai, China Automation Department, East China University of Science and Technology, Shanghai, China Automation Department, East China University of Science and Technology,Shanghai, China

会议类型: 国际会议

会议名称: 2011 4th International Congress on Image and Signal Processing(第四届图像与信号处理国际学术会议 CISP 2011)

会议地点: 上海

会议语种:英文

页码: 2039-2043

在线出版日期: 2011-10-15（万方平台首次上网日期，不代表论文的发表时间）