基于互学习的自适应PSO算法的亚像素定位研究

J4 ›› 2016, Vol. 38 ›› Issue (04): 747-754.

基于互学习的自适应PSO算法的亚像素定位研究

刘欢1，2,肖根福3，欧阳春娟1

（1.井冈山大学电子与信息工程学院，江西吉安 343009;
2.流域生态与地理环境监测国家测绘地理信息局重点实验室，江西吉安 343009;
3.井冈山大学机电工程学院，江西吉安 343009）

收稿日期:2015-05-21 修回日期:2015-07-10 出版日期:2016-04-25 发布日期:2016-04-25
基金资助:
国家自然科学基金(61462016)；江西省科技厅青年科学基金(20151BAB217012)；江西省科技厅自然科学基金(20151BAB207026)；江西省教育厅2014年度科技计划(GJJ14559)；江西省高校人文社科研究项目（YS1546）;流域生态与地理环境监测国家测绘地理信息局重点实验室资助课题；井冈山大学博士启动项目（JZB15016,JZB15009）

Subpixel registration based on adaptive particle
swarm optimization with mutual learning

LIU Huan1，2,XIAO Genfu3，OUYANG Chunjuan1

（1.School of Electronics and Information Engineering,Jinggangshan University,Ji’an 343009;
2.Key Laboratory of Watershed Ecology and Geographical Environment Monitoring,NASG,Ji’an 343009;
3.College of Machanical & Electrical Engineering,Jinggangshan University,Ji’an 343009,China）

Received:2015-05-21 Revised:2015-07-10 Online:2016-04-25 Published:2016-04-25

摘要/Abstract

摘要：

针对数字图像相关方法（DIC）的亚像素精确定位运算量大、时间代价高的问题，提出了一种改进的粒子群优化方法的亚像素精确定位。依据待测物图像中特征点变形程度的差异自适应地调整粒子飞行的速度和范围并细化到x和y二维方向上，改善特征点位移解的质量；另外，引入粒子间的互相学习机制，充分利用前一粒子的历史信息，减少迭代次数，提高算法运行效率；最后，将这种互学习的自适应粒子群的亚像素定位算法与牛顿拉夫森（NewtonRaphson）算法和牛顿拉夫森粒子群（NRPSO）算法作比较。实验结果表明，本文算法具有更高的精度、有效性和可行性，尤其在处理大数据量时，该算法的时间成本优势更为显著。

Abstract:

In order to solve the problem of huge computation and high time cost in subpixel registration resolution of digital image correlation, we propose a new improved PSO for subpixel registration of the respective flight velocities. The flying velocity and range of particles which are subdivided at two directions x and y, can adaptively adjust according to the deformation degree of each interest point so as to improve their displacement solution quality. In addition, the reliable mutual learning mechanism is introduced and the historical information of the previous feature points is fully utilized, which helps to reduce the number of iterations and enhance algorithm efficiency. Compared with the NewtonRaphson and the NRPSO, the proposed method has higher accuracy, and the feasibility and availability are verified. Particularly, the superiority of time cost is more distinct when dealing with a large number of interest points.

Key words: digital image correlation;subpixel registration resolution;mutual learning adaptive particle swarm optimization;NewtonRaphson algorithm

刘欢1，2,肖根福3，欧阳春娟1. 基于互学习的自适应PSO算法的亚像素定位研究[J]. J4, 2016, 38(04): 747-754.

LIU Huan1，2,XIAO Genfu3，OUYANG Chunjuan1. Subpixel registration based on adaptive particle
swarm optimization with mutual learning [J]. J4, 2016, 38(04): 747-754.