CAJ | 학술논문

针对基于内容的图像检索技术存在的问题,结合多维空间仿生信息学（Multi-Dimensional SpaceBiomimetic Informatics,MDSBI）理论和可并行计算原理,根据在多维空间中分析图像的同源连续性规律和GPU工作的特点,提取一种可并行计算的图像特征提取算法和检索技术.实验比较证明,该方法在解决检索效率和检索速度有很大的提高.
침대기우내용적도상검색기술존재적문제,결합다유공간방생신식학（Multi-Dimensional SpaceBiomimetic Informatics,MDSBI）이론화가병행계산원리,근거재다유공간중분석도상적동원련속성규률화GPU공작적특점,제취일충가병행계산적도상특정제취산법화검색기술.실험비교증명,해방법재해결검색효솔화검색속도유흔대적제고.
The paper dwells on the problems related to visual media retrieval on the basis o! contents. The discussion sets forth from the principle of homology continuity （PHC）of Multi-Dimensional Space Biomimetic Informatics （MDSBI）, analyzes videos, shots and images in multi-dimensional space, and attempts to examine the heterogeneity between different visual media in the perspective of subspace, thus arriving at the conclusion that visual media retrieval should first of all tackle the issue of analyzing the distance between dots and subspace and come up with an algorithm of shot segmentation for visual media retrieval.