光学技术, 2020, 46 (3): 354, 网络出版: 2020-07-16
基于人眼视觉敏感特性的图像清晰度评价
Image sharpness evaluation based on human visual sensitivity
图像质量评价 无参考清晰度评价 人眼视觉敏感特性 高通滤波 Image quality evaluation no reference sharpness evaluation human visual sensitivity high-pass filtering
摘要
提出一种结合人眼视觉敏感特性的无参考图像清晰度客观评价方法,利用高通滤波划分出人眼视觉敏感(图像细节)和非敏感区域,分别计算对应区域的清晰度,最后加权求和得出整幅图像的清晰度。利用公开的4个图像库与现有的6个无参考清晰度评价模型进行测试和比较,所研究的评价模型计算结果与主观评价分数有较好的一致性,且运算速度较快。
Abstract
A non-reference objective image sharpness evaluation method considering human visual sensitivity is proposed. A high-pass filtering is used to divide the human visually sensitive area that mainly contains image details and the non-human visually sensitive area. Then the sharpness of the two areas are calculated respectively. and its weighted sum is the final sharpness of the whole image. The published 4 image libraries to test and compare is used. with 6 existing no-reference sharpness evaluation models. The results show that the model is in good agreement with the subjective evaluation score and possesses faster calculation speed.
刘玉玺, 郑秋梅, 高绍姝, 张晓东, 张威. 基于人眼视觉敏感特性的图像清晰度评价[J]. 光学技术, 2020, 46(3): 354. LIU Yuxi, ZHENG Qiumei, GAO Shaoshu, ZHANG Xiaodong, ZHANG Wei. Image sharpness evaluation based on human visual sensitivity[J]. Optical Technique, 2020, 46(3): 354.