A ROBUST SUPER-RESOLUTION USING COMBINATION OF DIFFERENT TYPE FEATURE DETECTIONS
Recently, media for pictures and movies shift from analog data on films to digital data. These data is sub-sampled and have low resolution. Therefore, some techniques of conversion or estimation high resolution image from low resolution images are presented. These algorithms generating high resolution image are called as super-resolution. However, there are some problems, huge processing time, and necessity to use high performance processor. For resolving these problems, some improving researches are progressing, In this paper, an accurate registration technique using combination of different type feature detections is proposed as a quality raising technique. Registration, or position estimates, is used in one of the super-resolution techniques which use plural low resolution images to superpose these low resolution images.
super-resolution registration feature detection morphing
Hitoshi Yamauchi Akihiro Kanagawa
Faculty of Computer Science and Systems Engineering, Okayama Prefectural University,Okayama 719-1197, Japan
国际会议
The Tneth International Conference on Industrial Management(第十届工业管理国际会议 ICIM 2010)
北京
英文
729-734
2010-09-16(万方平台首次上网日期,不代表论文的发表时间)