摘要
目的针对现有图像匹配算法因忽略图像颜色信息,使算法对光照变化的鲁棒性较差,容易造成图像误匹配的情况进行改善。方法颜色信息是区分不同图像的重要特征,通过提取彩色图像颜色不变量的方法获得图像的颜色信息,并将颜色不变量与ORB特征描述子相结合,形成一种C-ORB算法,C-ORB算法在保留ORB算法优点的基础上,能够将图像的色彩信息应用到图像匹配过程当中。结果实验结果表明,结合了颜色不变量的ORB算法在保留了ORB算法运算速度快的基础上,能够在不同光照强度、不同光照方向下,获得更多的图像细节,匹配更为准确。结论在移动手机、平板电脑等移动终端设备进行印刷品识别与匹配时,能够在保证识别图像识别速度的基础上进一步减少图像的误匹配。
This paper aims to solve the problem with the current image matching algorithm which ignores image information, thus leads to bad robustness of illumination variation and easily causes error matching. Color is an important component for distinction between images. In this paper, we presented a new ap proach based on color invariant and ORB feature, called C-ORB. It applied color information to the image matching while retaining the ORB ＇s efficiency and performance. Experimental results showed that C-ORB retained the ORB＇s efficiency, could get more image details, and was more accurate and stable with respect to variations in the photometrical imaging conditions. In conclusion, for recognition and matching of pressworks on mobile terminals like mobile phone and tablet PC, this algorithm can further re duce error matching while retaining the original speed of image recognition.
出处
《包装工程》
CAS
CSCD
北大核心
2016年第15期18-23,共6页
Packaging Engineering
