摘要
针对实时视频图像拼接系统对拼接图像进行客观质量评估的需求,在图像质量评价方法的基础上,根据影响拼接质量的两大因素(重影和亮度差异),提出一种基于图像差分信息熵的拼接图像质量评估方法。方法首先对原图像和拼接图像进行边缘提取,然后利用图像的边缘轮廓信息进行差分处理,得到边缘差分图,再根据边缘差分图的信息熵值与影响拼接两大因素的关系,对拼接图像评估。最后通过仿真,与基于结构相似度算法对比,证明所提方法得出的评分结果更加符合人眼视觉对拼接图像的主观评价。
In order to meet the requirement of real-time video image mosaic system for objective quality assessment of mosaic images, a method of image stitching quality assessment based on differential image entropy is proposed according to the two main factors(geometric dislocation and luminance difference) which affect stitching quality. Firstly, this method extracted the edge of the original image and mosaic image. Then using the edge contour information of the image to carry on the differential processing, the edge difference map was obtained. According to the relationship between the information entropy of the edge differential diagram and the two factors affecting stitching, the stitching image was evaluated. Finally, the simulation results show that the proposed method is more consistent with the subjective evaluation of mosaic images by human vision than SSIM.
作者
王林
王超凡
WANG Lin;WANG Chan-fan(College of Automation and Information Engineering,Xi’an University of Technology,Xi’an Shanxi 710048,China)
出处
《计算机仿真》
北大核心
2020年第4期265-268,273,共5页
Computer Simulation
基金
陕西省科技计划重点项目(2017ZDCXL-GY-05-03)。
关键词
图像处理
图像拼接
客观质量评估
差分信息熵
Image processing
Image stitching
Objective quality evaluation
Information entropy of differential graph