基于BP神经网络的图像感兴趣区自动检测技术  

Automatic detecting region of interesting technique based on BP neural network

作　　者：尹显东[1] 姚军[2] 李在铭[1] 

机构地区：[1]电子科技大学通信学院,四川成都610054 [2]中国工程物理研究院电子工程研究所,四川绵阳621900

出　　处：《系统工程与电子技术》2006年第2期192-195,287,共5页Systems Engineering and Electronics

基　　金：总装备部"十五"预研基金(30020490523);NSAF联合基金(10376005);中国工程物理研究院电子工程研究所创新基金(S20041111)资助课题

摘　　要：图像中感兴趣区(region of interesting,ROI)的自动分割是图像信息压缩与模式识别中的重要问题。提出一种图像中感兴趣区的图像信息自动检测算法,通过建立基于BP神经网络的多层检测模型,自动实现图像中ROI的自动检测与分割。并将该模型用于对车牌图像中牌照区域图像的自动分割。结果表明,该算法稳定性好,能够方便快速地实现特定ROI的自动检测与分割。The technology based on region of interesting （ROI） has already become the important research field of computer image processing, and it has been applied effectively in the image coding compressing and image recognition. A adaptive method is introduced to detect and extract the ROI, a multi-layer detect model is developed based on BP neural network to detect automatic and segment ROI form the image with complex background. Using license plate image makes practical detecting and segment experiments. The results show that the method is an efficient, quick and stabilization detecting algorithm for detecting and segment ROI form image with complex background, And it has great value in computer image processing,

关 键 词：图像分割 图像通信 BP神经网络 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术] TP309.2[自动化与计算机技术—计算机科学与技术]