摘要
为了提高图像边缘检测的细节信息,采用了二进制粒子群算法。首先通过logistic变换更新粒子速度,粒子速度不受限制;接着动态调整粒子位置,使飞行状态充分调整;然后正态云发生器动态调整粒子的惯性权重,这样较优粒子可以取得较小的惯性权重;最后建立图像边缘检测模型和算法流程。实验仿真显示本文算法对图像边缘定位准确、清晰,信噪比为35.928 1db,处理时间为1.340 1s。满足检测结果中对信息含量大、执行时间少等要求。
The presented binary particle swarm optimization algorithm is used to improve the quality of image edge detection. First, the particle velocity is updated through logistic tranform which isn＇t re-stricted. Second, position of particle is adjusted dynamically, and state of flight is adjusted fully. Third, inertia weight of particle is adjusted dynamically by normal cloud generator, smaller particle is got optimum inertia weight. Finally, the processing of image edge detection is described. The experiment results show that the edge detection is precisely and clear,SNR is 35. 928 ldb, time is 1. 454 3 s.It has large information and takes less time.
出处
《液晶与显示》
CAS
CSCD
北大核心
2014年第5期800-804,共5页
Chinese Journal of Liquid Crystals and Displays
基金
河南省政府重大专项(No.121100111000)
关键词
粒子群
二进制
边缘检测
灰度
particle swarm optimization
binary
edge detection
gray
