数学形态学在储粮昆虫图像识别中的应用  

APPLICATION OF MATHMETICS PORPHILOGY ON THE IMAGE SEGMENTATION AND MODELRECOGNITION OF STORED FOOD BEETLE

作　　者：张学庆[1] 刘燕[2] 

机构地区：[1]中国人民解放军通信指挥学院,湖北武汉430010 [2]华中光电技术研究所,湖北武汉430073

出　　处：《武汉工业学院学报》2004年第1期94-96,共3页Journal of Wuhan Polytechnic University

摘　　要：充分分析了储粮昆虫图像分割在图像模式识别中的必要性,然后提出一种基于数学形态学的改进的图像分割方法。与一般的将物体从背景中分割出来的分割方法不同,这种基于数学形态学的改进的图像分割方法,不仅可以将储粮昆虫的足和触角分割出来,而且很好的保留了各个部分的边缘细节特征。In this paper we analyzed the necessity of image segmentation in stored food beetle recognition. An image segmentation method based on mathematical morphology is put forward, which is different from any other way of segmenting objects from the background. This method can segment the beetle's feet and antennas from the body and reserves every part's characters in detail, especially on the edge. The results of the test with this method show the effectiveness of this method in segmenting stored food beetle's feet and antennas from the body.

关 键 词：储粮昆虫 图像模式识别 数学形态学 图像分割 

分 类 号：O1[理学—数学]