摘要
针对用传统的颜色传递算法得到的彩色融合图像存在目标与背景颜色相近、目标变淡的问题,提出了一种基于颜色对比度增强的红外与微光图像彩色融合方法。首先对红外和微光图像进行了伪彩色融合,然后对彩色参考图像与伪彩色融合图像进行了颜色传递。其次,采用大津分割法分割出了微光图像中的目标信息,并得到了目标区域。最后,在HSI颜色空间中利用目标区域对颜色传递后的融合图像的H、S、I分量进行了调整,并得到了最终的融合图像。实验结果表明,利用该方法得到的彩色融合图像具有目标与背景对比度高、细节丰富、色彩较好等特点,可提高人眼对场景的理解和对目标的快速识别。
In the color fusion images obtained by traditional color transfer algorithms, the color of the target is similar to that of the background and the target becomes faded. To solve this problem, an infrared and low-level-light image color fusion method based on color contrast enhancement is proposed. Firstly, the infrared image and the low-level-light image are fused in color. Then, the color transfer is implemented on the color reference image and the pseudocolor fusion image. Secondly, the Otsu segmentation method is used to extract the target information in the low-level-light image and hence the target area image is obtained. Finally, the target area in the HIS color space is used to adjust the HIS component of the fusion image obtained through color transfer. The final fused image is obtained. The experimental result shows that the color fusion image obtained by this method has the features of high target-background contrast, rich detail and better color. It can help human eyes detect and identify targets quickly.
出处
《红外》
CAS
2016年第11期14-17,23,共5页
Infrared
关键词
伪彩色融合
颜色传递
大津分割
红外
微光
color fusion
color transfer
Otsu segmentation
infrared
low-level-light
