Research On Concrete Lining Seams Elimination Of Tunnel Surface Image Based On Line Segment Extraction

Highway tunnel is an important part of highway traffic infrastructure,crack is a common and harmful disease.Timely and effective detection of tunnel cracks is of great significance to the later maintenance of the tunnel.The crack detection technology based on digital image processing overcomes the shortcomings of traditional manual detection and instrument detection and lays a foundation for the daily maintenance and hazard rating of the tunnel.In addition,due to the similar shape of highway tunnel cracks and inherent lining joints,the lining joint removal algorithm has a broad application prospect in the field of crack detection,which has very important theoretical significance and practical value.How to accurately detect and remove the lining joints on the tunnel lining surface in the image with cracks is researched in this thesis,the technology status of lining joint removal and crack detection at home and abroad are summarized,the existing line detection technology is used to extract the significant line segment edge,and the line segment edge are classified,and the lining joint removal algorithm based on rectangular mark are puts forwarded to complete the lining joint removal and ensure the safety.The main research work is as follows:1.Aiming at the problem of large performance difference of existing line detection algorithms,the current methods of line segment extraction is analyzed and the line segment coarse detection and significant line segment extraction for tunnel lining surface image are carried out in this thesis.By analyzing the performance index of the results of line detection algorithm,LSD(line segment detector)line detection algorithm is used as the line extraction algorithm in this thesis to extract significant lines,which provides a reliable basis for the subsequent experiments.2.Because the characteristics of lining joints and cracks are very similar,the existing distinguishing methods are not accurate enough.By analyzing the distribution characteristics and differences of cracks and lining cracks,an adaptive threshold setting method based on angle adaptation is designed,the comprehensive performance index of linear angle and length of lining cracks and lining cracks are studied,and the lining cracks are distinguished effectively from lining cracks in this thesis.At the same time,compared with the existing judgment methods,the effectiveness of the adaptive threshold method is verified,and the possible areas of lining joints are found,so as to improve the efficiency and accuracy of the subsequent lining joint removal algorithm.3.In view of the unpredictable direction of lining joints and cracks,the existing methods have low removal accuracy for lining joints in complex directions.How to remove the lining joint quickly and effectively in the tunnel lining surface image is studied and a lining joint removal algorithm based on rectangular marking is proposed in this thesis.The edges of the straight line segments in the threshold area are marked with rectangular marks one by one,which are the reserved area and the removed area.Finally,according to the marking information,the pixels are retained or removed to remove the lining joint accuratelyFinally,the research work is summarized and the future research direction is prospected in this thesis.