计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (16): 210-214.

• 图形图像处理 • 上一篇    下一篇

近距大偏角圆形目标的检测

刘晓俊,孙永荣,王潇潇,杨博文   

  1. 南京航空航天大学 自动化学院 导航研究中心,南京 210016
  • 出版日期:2015-08-15 发布日期:2015-08-14

Detection of close ranged circular target on large deflection angle

LIU Xiaojun, SUN Yongrong, WANG Xiaoxiao, YANG Bowen   

  1. Navigation Research Center, College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Online:2015-08-15 Published:2015-08-14

摘要: 圆形目标距离图像采集设备较近且存在较大偏转角时成像会产生严重的形变,给目标图像的自动检测带来了困难。提出了一种近距大偏角圆形目标的检测算法,利用闭运算处理,去除图像背景中的噪声,并采用嵌套双层轮廓提取方法实现目标轮廓提取,通过轮廓特征提取算法获得目标轮廓的初选轮廓,利用初选轮廓的细节特征信息确认目标轮廓。为了实现目标的高效检测,采用模板匹配算法进一步提高了算法运行效率以及稳定性。实验结果表明,所提出的检测算法可以有效地适应不同的背景环境,具有较强的鲁棒性和实时性。

关键词: 轮廓提取, 模板匹配, 圆形目标, 阈值分割, 闭运算

Abstract: Serious deformation of the circular target can be caused by the existence of short distance and large deflection angle between the circular target and the image acquisition device. It will bring difficulty to the automatic detection of the target. A detection algorithm of close ranged circular target is figured?out. The close operation treatment is used for de-noising the background of images. The contour can be extracted by the extraction algorithm of nested double contour. Original contours can be got from all the contours by the contour feature extraction algorithm. The target contour is confirmed by detailed features of the original contours. Besides, the efficient detection of the target is implemented by adopting template matching algorithm to improve the efficiency and stability of the algorithm. The detection algorithm has been proven that it can effectively adapt to different environmental background and is very robust and real-time.

Key words: contour extraction, template match, circular target, threshold segmentation, closed operation