Airport Runway Glue Mark Detection System Based on Image Processing Technology
Aiming at the problems of low efficiency and insufficient precision of the current airport runway glue mark detection method,an automatic detection system for airport runway glue marks based on computer vision is de-signed.Firstly,a glue mark image filtering algorithm based on noise judgment rules is adopted by using an adaptive threshold and sliding window to improve the denoising ability;then,pinhole imaging and Metropolis criterion are in-troduced to improve the searchability and convergence speed of the sparrow search algorithm,and the improved spar-row search algorithm was used to optimize K-Means for glue mark segmentation to improve the segmentation accuracy.Finally,a physical platform was built to verify the feasibility and effectiveness of the system.The experimental results show that when dealing with multi-dimensional functions,the convergence accuracy and speed of the improved sparrow search algorithm are improved by 0.6 and 5 orders of magnitude respectively;the SSIM of the denoised glue mark image can reach 0.7;the improved image segmentation algorithm can accurately segment the runway glue marks.The system has good adaptability and anti-interference.
System designAirport runway glue marksImage denoisingSparrow search algorithmImage seg-mentation
NETL
NSTL
万方数据
