安阳工学院学报
安暘工學院學報
안양공학원학보
JOURNAL OF ANYANG INSTITUTE OF TECHNOLOGY
2013年
6期
47-49,110
,共4页
陈姝颖%陈雷%杜萍%赵小芳
陳姝穎%陳雷%杜萍%趙小芳
진주영%진뢰%두평%조소방
图像检索%颜色直方图%图像熵%空间信息%重叠分块
圖像檢索%顏色直方圖%圖像熵%空間信息%重疊分塊
도상검색%안색직방도%도상적%공간신식%중첩분괴
image retrieval%color histogram%image entropy%spatial information%block overlapped
针对利用图像的颜色特征丢失空间信息的问题,提出了一种综合方法:通过计算图像熵,反映图像中像素邻域内颜色分布的综合特征;通过重叠分块,然后对各分块计算分块颜色直方图。实验结果表明,这种方法具有很好的检索效率。
針對利用圖像的顏色特徵丟失空間信息的問題,提齣瞭一種綜閤方法:通過計算圖像熵,反映圖像中像素鄰域內顏色分佈的綜閤特徵;通過重疊分塊,然後對各分塊計算分塊顏色直方圖。實驗結果錶明,這種方法具有很好的檢索效率。
침대이용도상적안색특정주실공간신식적문제,제출료일충종합방법:통과계산도상적,반영도상중상소린역내안색분포적종합특정;통과중첩분괴,연후대각분괴계산분괴안색직방도。실험결과표명,저충방법구유흔호적검색효솔。
This paper puts forward a comprehensive method because of the lost of the spatial information when using the color feature of the image: it can reflect the comprehensive features of the color distribution in the pixel neighborhood to calculate the image entropy, we block overlapped the image then calculate the block-color histogram. We retrieve the image integrated the above two methods. The experimental results show that, this method has very good retrieval efficiency.