Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (28): 73-74.

• 学术探讨 • Previous Articles     Next Articles

Method for recognizing colored target based on improved SSDA

SUN Li-gong1,2,RAO Wen-bi1,YAN Bao-ding2   

  1. 1.School of Computer Science and Technology,Wuhan University of Technology,Wuhan 430070,China
    2.Electronic Information Engineering College,Henan University of Science and Technology,Luoyang,Henan 471003,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-10-01 Published:2007-10-01
  • Contact: SUN Li-gong

基于改进SSDA的彩色目标识别方法

孙立功1,2,饶文碧1,阎保定2   

  1. 1.武汉理工大学 计算机学院,武汉 430070
    2.河南科技大学 电子信息工程学院,河南 洛阳 471003
  • 通讯作者: 孙立功

Abstract: A method based on improved SSDA(Sequence Similarity Detection Algorithm) is proposed,it is applied in recognizing the colored target of robotic visual.SSDA is improved based on color weight coefficient.Furthermore,the information of shape and size is introduced while extracting image characteristics.The experiment results show that the algorithm reduces the amount of calculation and increases the accuracy of recognizing target.Therefore,the algorithm has well real-time performance and robustness.

Key words: image matching, Sequence Similarity Detection Algorithm(SSDA), accumulative errors, weight coefficient

摘要: 提出了基于改进SSDA算法的机器人视觉彩色目标识别方法,利用颜色分量的权重系数对SSDA算法进行了改进,同时,在图像特征提取时引入目标的形状和大小信息。实验表明,这些措施有效地减少了运算量,提高了目标识别的准确性,具有较好的实时性和鲁棒性。

关键词: 图像匹配, 序贯相似性检测算法, 累计误差, 权重系数