An image overall complexity evaluation method based on LSD line detection

3rd International Conference on Energy Materials and Environment Engineering

能源学;生态环境科学

Li, Jianan^1,2,3 ; Duan, Jin^1,2,3 ; Yang, Xu^1,2,3 ; Xiao, Bo^1,2,3

School of Electronic and Information Engineering, Changchun University of Science and Technolog, Changchun, China^1

Space Photoelectric Technology Research Institute, Changchun University of Science and Technology, Changchun, China^2

Changchun University of Science and Technology, Changchun, China^3

关键词: Artificial worlds;    Complexity calculation;    Degree of dispersion;    Expert consultation;    Line detection algorithms;    Linear information;    Straight line detection;    Weight coefficients;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/61/1/012162/pdf
DOI  :  10.1088/1755-1315/61/1/012162

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

In the artificial world, whether it is the city's traffic roads or engineering buildings contain a lot of linear features. Therefore, the research on the image complexity of linear information has become an important research direction in digital image processing field. This paper, by detecting the straight line information in the image and using the straight line as the parameter index, establishing the quantitative and accurate mathematics relationship. In this paper, we use LSD line detection algorithm which has good straight-line detection effect to detect the straight line, and divide the detected line by the expert consultation strategy. Then we use the neural network to carry on the weight training and get the weight coefficient of the index. The image complexity is calculated by the complexity calculation model. The experimental results show that the proposed method is effective. The number of straight lines in the image, the degree of dispersion, uniformity and so on will affect the complexity of the image.