用于精确定位的最佳匹配区选择分形法  

Selection of Suitable Matching Area by Fractal Based Approach for High Precision Location

作　　者：李俊[1] 杨新[1] 朱菊华[1] 施鹏飞[1] 

机构地区：[1]上海交通大学图像处理与模式识别研究所,上海200030

出　　处：《上海交通大学学报》2001年第2期305-308,共4页Journal of Shanghai Jiaotong University

基　　金：国家自然科学基金!资助项目 (6 0 0 72 0 2 6 )

摘　　要：图像中任意点局部邻域的分形维数越大 ,该邻域图像数据的相关性越小 ,相关匹配搜索时定位该区域就越容易 .由此提出了利用分形维数作为分离参数 ,在基准地图上选择最佳匹配区的方法 .用相关法对选出的最佳匹配区验证表明 ,分形方法比传统的相关方法在选择最佳匹配区时 ,定位精度更高 ,计算更快 .In the scene matching based robot or flight vehicle navigation system, the pre-loaded reference images are compared with the real-time acquisition images by correlation method for position locating. And it is important to choose most distinguishable scenes as the suitable-matching area (SMA) for high precision location. This paper addressed the close relation between the fractal dimension (FD) and the correlate coefficient of the matched image, that is, the greater the FD is, the less relative the image data. Thus, it suggested a method based on FD of the image to select SMA from the reference image, which is critical for navigation. The experiments prove that the proposed method has the advantages of accurate and reliable extraction of the SMA from the reference images, and high computational efficiency.

关 键 词：地形匹配 相关匹配算法 分形 最佳匹配区 机器人 定位 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]