基于粒子群优化算法的水果检测算法  

Circle Detection Based on Particle Swarm Optimization

作　　者：徐辉 袁彪 王维俊 邹益民[1] 刘静[1] 朱霞[1] Xu Hui;Yuan Biao;Wang Weijun;Zou Yimin;Liu Jing;Zhu Xia(Jinling Institute of Technology,Nanjing Jiangsu,211169)

机构地区：[1]金陵科技学院

出　　处：《电子测试》2019年第15期49-52,59,共5页Electronic Test

基　　金：江苏省高等学校自然科学研究面上项目(No.17KJB510020)

摘　　要：针对霍夫变换进行圆检测时计算量大、耗时长等问题,提出了一种基于改进型粒子群的圆检测算法。该算法先对拍摄图像进行边缘检测获取边缘图像后,再从边缘点中随机选取1点作为初始粒子位置,通过设置最大算法迭代次数与阈值来克服粒子群算法陷入局部最优问题。以每个粒子作为一个候选圆心,采用优化策略,以粒子运动、融合更新等操作完成圆的检测。A circle detection algorithm was proposed based on particle swarm optimization in order to remove some problems of Hough transform which has a large amount of calculation and cost a lot of time. The edge image was obtained through pre-process grayscale, filtering noise and edge detection, one point was selected randomly as the initial particle position. In the process of circle detection, the maximum iterations and the threshold were set in order to not fall into local optimum. Each particle was taken as a candidate circle, then used the optimization strategy, circle detection was completed with the particle motion and update operation.

关 键 词：粒子群算法 霍夫变换 圆检测 惯性因子 收敛因子 

分 类 号：O15[理学—数学]