摘要
为了解决激光切割视觉识别系统对大幅面多目标图像识别实时性差、对带有偏角的目标识别率低甚至出现无法识别等问题,采用小生境遗传算法,对大幅面多目标匹配识别算法进行了理论分析和实验验证。结果表明,无论目标有无旋转,该算法都能达到100%识别,算法运算时间比传统算法快5倍~10倍。该算法在识别带有旋转角度的多目标时,具有很好的实用性。
In order to solve the problem that the real-time performance of the laser cutting visual identity system was poor in the recognition of large and multi-target images,and the recognition rate of targets with deflection angle was low or even zero,niche genetic algorithm was used to study the large-surface multi-target matching recognition algorithm.Theoretical analysis and experimental verification were carried out.The results show that,whether the target is rotating or not,the algorithm can reach recognition rate of 100%.The computation time of the algorithm is 5 times~10 times faster than that of the traditional algorithm.The algorithm has good practicability in identifying multiple targets with rotation angle.
引文
[1]HU M X,WANG X L.Template matching optimization coupled image correction for rotating workpiece target localization algorithm[J].Modular Machine Tool and Automated Processing Technology,2016(6):35-38(in Chinese).
[2]SHEN X,GUO H,WANG Y.Workpiece assembly reference end shape recognition research[J].Modern Manufacturing Engineering,2015(2):97-101(in Chinese).
[3]ZHANG Zh F,ZHANG Y R,PEI Zh L.OpenCV coupling humancomputer interaction of mobile phone surface target detection and positioning research[J].Modular Machine Tools and automated Processing Technology,2015(3):67-70(in Chinese).
[4]ZHANG Y R,PEI Zh L.Open CV coupling three-eye vision of standard parts target positioning research and application[J].Modular Machine Tools and Automated Processing Technology,2015(1):67-70(in Chinese).
[5]CHEN Z N.Research on workpiece location based on template matching[D].Guangzhou:Guangdong University of Technology,2016:1-4(in Chinese).
[6]ZHANG X.Research on the application of vision positioning technology in laser cutting equipment[D].Guangzhou:Guangdong University of Technology,2016:2-5(in Chinese).
[7]SHU Zh G.Research on license plate image tilt correction algorithm based on DSP[D].Hefei:Anhui University of Technology,2014:15-20(in Chinese).
[8]XU P F.Application and implementation of bilinear interpolation and cubic convolution in image scaling[J].Network Security Technology and Application,2017(12):51(in Chinese).
[9]HU Y W.Application of genetic algorithm in image recognition[D].Wuhan:Wuhan University of Technology,2006:61-63(in Chinese).
[10]LI Y,GUI W X.An image segmentation method based on improved niche genetic algorithm[J].Computer Applications and Software,2017,34(4):202-206(in Chinese).
[11]FEI X Y,WANG F.An improved niche genetic algorithm[J].Journal of Chongqing Post and Telecommunications University(Natural Science Edition),2005,17(6):721-723(in Chinese).
[12]ZHANG T F,ZHANG H X,MENG F,et al.Application of improved Hausdorff distance and quantum genetic algorithm in laser image guidance[J].Laser Technology,2016,40(3):320-325(in Chinese).
[13]LING Zh J,WANG X P,XUE X P.Application of improved niche genetic algorithm in multi-objective optimization[J].Computer Applications and Software,2008,25(11):16-19(in Chinese).
[14]XUE J.Research on image matching method based on improved genetic algorithm[D].Taiyuan:Taiyuan University of Technology,2011:19-23(in Chinese).
[15]SHEN Y Ch,TANG W W,ZHANG G X,et al.Research of routing problem based on genetic algorithm[J].Laser Technology,2011,35(3):422-424(in Chinese).
[16]LIANG P,LIU M Zh.A niche genetic algorithm for multi-chromosome crossover[J].Computer Engineering and Applications,2016,52(18):162-166(in Chinese).
[17]YANG Y,HUANG Zh Q,LIU F Y,et al.Application of multi-population genetic algorithm in wet steam parameter measurement[J].Laser Technology,2014,38(4):484-487(in Chinese).