CAJ | 학술논문

针对传统的基于文本关键字的图像检索方法已经不能满足对大型医学数据库检索的需要的问题，本文提出了一种基于高维特征检索智能算法的图像检索技术。该算法使用高维向量设计基于内容的图像检索，建立训练样本集，根据样本集的距离设置遗传算法的氏适应度函数，而后遗传迭代出最优向量组合，进行测试集的匹配和检索。算法仿真结果表明，本文提出的基于高维特征检索智能算法的图像检索技术在检索精度上比传统方法高。
침대전통적기우문본관건자적도상검색방법이경불능만족대대형의학수거고검색적수요적문제，본문제출료일충기우고유특정검색지능산법적도상검색기술。해산법사용고유향량설계기우내용적도상검색，건립훈련양본집，근거양본집적거리설치유전산법적씨괄응도함수，이후유전질대출최우향량조합，진행측시집적필배화검색。산법방진결과표명，본문제출적기우고유특정검색지능산법적도상검색기술재검색정도상비전통방법고。
In view of the traditional image retrieval methods based on keyword text already cannot satisfy the need for large medical database retrieval problem, this paper proposed an intelligent algorithm based on high dimensional feature retrieval of image retrieval technology. The content based image retrieval algorithm using the high-dimensional vector design, establish the training sample set, according to the distance of the sample set set's fitness function of genetic algorithm, and then genetic iterative optimal vector combination, test sets of matching and retrieval. Algorithm simulation results show that the proposed intelligent algorithm of image retrieval technology based on high dimensional feature retrieval on the retrieval accuracy is higher than traditional methods.