路标图像识别的禁忌搜索遗传算法研究  

Research on genetic algorithm based on tabu search for landmark image recognition

作　　者：吴晓燕[1] 

机构地区：[1]四川文理学院计算机学院,四川达州635000

出　　处：《计算机工程与设计》2014年第6期2109-2113,共5页Computer Engineering and Design

基　　金：四川省教育厅2014年度一般基金项目(14ZB0315)

摘　　要：针对移动智能机器人基于单目视觉的路标识别问题,提出了一种融合色彩信息并基于禁忌遗传搜索算法的路标图像提取算法,并给出了收敛性分析。进行RGB视觉模型的实验闭值图像分割,经边缘提取后,采用禁忌搜索的遗传算法,提取路标图像的基本形状,得到待识别路标轮廓;将待识别路标的信息反馈到灰度图进行二次分割,完成定位和提取;利用Hopfield神经网络对归一化的路标进行分类,完成最终识别。对多幅小同场景图像进行了实验,实验结果表明,该方法能够快速有效地从图像中自动定位、提取与识别路标。To solve the problem of monocular-vision-oriented landmark recognition for mobile intelligent robot marching with complex background,a kind of landmark extraction algorithm which fused with transcendental color information based on tabugenetic search algorithm was proposed,and the convergence analysis of this algorithm was presented.Firstly,a novel experiment threshold based on RGB vision model was used for the detection step on the target image,after extracting image edge,the tabu-genetic search algorithm started to work for recognizing landmark image primitive shape （primitive element）,and so landmark contours were extracted for next position and recognition steps.Then,the position information as feedback was conveyed to the grayscale image,the whole landmark was extracted accurately with the second segmentation on the local image that was exclusive to landmark area.Finally,landmarks were recognized by Hopfield neural network Results issued from experiments on a great number of images in deferent scenes showed that a satisfied recognition effect had reached that landmarks could be recognized fast and efficiently.

关 键 词：禁忌搜索 遗传算法 路标识别 神经网络 收敛性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]