湖南工程学院学报:自然科学版
湖南工程學院學報:自然科學版
호남공정학원학보:자연과학판
Journal of Hunan Institute of Engineering(Natural Science Edition)
2012年
1期
20-23
,共4页
图像拼接%大米%检测系统
圖像拼接%大米%檢測繫統
도상병접%대미%검측계통
image stitching%rice%detection system
图像拼接技术主要包括图像匹配和图像融合,根据大米检测系统获取图像的特性,在块模板匹配算法基础上提出了一种改进的块模板匹配算法.并通过大米图像实例把两种算法进行了比较分析,实验结果表明改进的块模板匹配算法效率更高,匹配效果更好.采用了最为常用的加权平均法中的渐入渐出的图像融合算法.
圖像拼接技術主要包括圖像匹配和圖像融閤,根據大米檢測繫統穫取圖像的特性,在塊模闆匹配算法基礎上提齣瞭一種改進的塊模闆匹配算法.併通過大米圖像實例把兩種算法進行瞭比較分析,實驗結果錶明改進的塊模闆匹配算法效率更高,匹配效果更好.採用瞭最為常用的加權平均法中的漸入漸齣的圖像融閤算法.
도상병접기술주요포괄도상필배화도상융합,근거대미검측계통획취도상적특성,재괴모판필배산법기출상제출료일충개진적괴모판필배산법.병통과대미도상실례파량충산법진행료비교분석,실험결과표명개진적괴모판필배산법효솔경고,필배효과경호.채용료최위상용적가권평균법중적점입점출적도상융합산법.
Image stitching technology includes image matching and image fusion. Acording to the image fea- tures of the rice detection system, this paper proposes an improved block template matching algorithm based on block template matching algorithm. Two algorithms are compared and analysed by the instance of rice image. Experimental results show that the improved block template matching algorithm is more effi- cient and better. In this paper, The gradually in and gradually out image fusion algorithm from weighted average algorithm is used.