An improved automatic image sequencing algorithm
Juan Wang · Journal of Optoelectronics·laser · 2009
An improved automatic image sequencing algorithm based on phase correlation is proposed.Firstly,the normalized cross-correlation function is computed,so the peak's value and position of the normalized cross-correlation function can be found.Then,the peak's value is used to estimate whether two images are overlapped.The head image and the last image can be ascertained automatically by maximal correlation degree intersecting.Finally,the relationship of the overlapped images can be ascertained by the peak's position.Experimental results show that the proposed method outperforms traditional methods,and it is of great value in image mosaics fields,which can sort the unordered image sequences effectively and automatically without manual work.