A novel regular-image mosaic method

作者：Chen Fushan; Sun Yuqiu^*; Li Xiong

来源：Journal of Information and Computational Science, 2014, 11(13): 4853-4859.

DOI：10.12733/jics20104440

• Characteristics
• City-block distances
• Distance functions
• Euclidean distance
• Image mosaic
• Nearest neighbor decision rules
• Nearest neighbor rule
• Reference image

This paper presents a novel regular image mosaic method. Firstly, select the reference images which have the same characteristic. Then, calculate the mean Euclidean distance (or the mean city-block distance) between the reference images and the rest images. Besides, the least Euclidean distance (or the least mean city-block distance) can be found out, as well as the image corresponding to the minimum. Then, put the two images together and check the rationality of the mosaic process according to the nearest neighbor decision rule based on the distance function. Next, repeat the above operations until the final image can be obtained. At last, a experiment has been done to check the rationality of this method. The experimental results show that this image mosaic method is suitable for regular images. This novel image mosaic process is highly-efficient and its accuracy is high.

• 出版日期2014