首页 | 本学科首页   官方微博 | 高级检索  
     

矩形和(椭)圆区域目标的分类识别
引用本文:朱殿尧,卞红雨. 矩形和(椭)圆区域目标的分类识别[J]. 激光与红外, 2009, 39(11): 1228-1232
作者姓名:朱殿尧  卞红雨
作者单位:1. 华北光电技术研究所,北京,100015;哈尔滨工程大学水声工程学院,黑龙江,哈尔滨,150001
2. 哈尔滨工程大学水声工程学院,黑龙江,哈尔滨,150001
摘    要:研究了矩形与(椭)圆形区域目标的分类识别问题,详细阐述了区分矩形和(椭)圆形区域目标的方法和步骤.在二值图像中存在着不同角度和尺寸的矩形与(椭)圆形区域目标,通过两种目标的面积、周长与其边界框对应参数的关系很好地区分了矩形和(椭)圆形目标,并给出目标的中心位置、长度(或长轴长)、宽度(或短轴长)、周长、面积及最小惯性轴的角度等重要几何参数.该方法直观,且具有旋转、平移、尺度不变性.

关 键 词:图像处理  目标分类  矩形  椭圆  面积  周长

Classification between rectangular and ellipsoid/circular areas
ZHU Dian-yao,BIAN Hong-yu. Classification between rectangular and ellipsoid/circular areas[J]. Laser & Infrared, 2009, 39(11): 1228-1232
Authors:ZHU Dian-yao  BIAN Hong-yu
Affiliation:North China Research Institute of Electro-optics,Beijing 100015,China;The Underwater Acoustic Engineering School of Harbin Engineering University,Harbin 150001,China
Abstract:The classification between rectangular and ellipsoid/circular areas was studied,and the method and steps was listed in detail.There were rectangular and ellipsoid/circular areas which had different areas and different obliquities in bio-value image.The rectangular areas and ellipsoid/circular areas targets could be classified by the parameters such as area,perimeter length and bounding box.And some important parameters such as the center position,length(major axis),width(minor axis),perimeter length,area,and the obliquity of the minor inertia axis were also given.The new method proposed in this paper was obvious and have rotation invariability and translation invariability.
Keywords:image processing  target classification  rectangular  ellipse  area  perimeter length
本文献已被 万方数据 等数据库收录!
点击此处可从《激光与红外》浏览原始摘要信息
点击此处可从《激光与红外》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号