山东大学学报(工学版)
山東大學學報(工學版)
산동대학학보(공학판)
JOURNAL OF SHANDONG UNIVERSITY(ENGINEERING SCIENCE)
2014年
3期
22-28
,共7页
兴趣区域%特征融合%语义关联%图像检索%半监督学习%正反馈
興趣區域%特徵融閤%語義關聯%圖像檢索%半鑑督學習%正反饋
흥취구역%특정융합%어의관련%도상검색%반감독학습%정반궤
region of interest%feature fusion%semantic correlation%image retrieval%semi-supervised learning%positive feedback
提出一种融合底层特征、基于兴趣区域的半监督学习图像检索方法,实现了图像内容的语义关联。该方法首先划分图像兴趣区域,提取图像的综合底层特征,然后将其作为训练数据,对图像类别进行半监督学习,建立图像和类别的语义映射,最后分别采用二次式距离和改进的 Canberra 距离对图像底层特征进行度量,特征空间中图像类的区域中心用正反馈进行迭代更新。通过实验对比,该图像检索算法具有较高的准确率,优于传统的基于内容的图像检索算法。
提齣一種融閤底層特徵、基于興趣區域的半鑑督學習圖像檢索方法,實現瞭圖像內容的語義關聯。該方法首先劃分圖像興趣區域,提取圖像的綜閤底層特徵,然後將其作為訓練數據,對圖像類彆進行半鑑督學習,建立圖像和類彆的語義映射,最後分彆採用二次式距離和改進的 Canberra 距離對圖像底層特徵進行度量,特徵空間中圖像類的區域中心用正反饋進行迭代更新。通過實驗對比,該圖像檢索算法具有較高的準確率,優于傳統的基于內容的圖像檢索算法。
제출일충융합저층특정、기우흥취구역적반감독학습도상검색방법,실현료도상내용적어의관련。해방법수선화분도상흥취구역,제취도상적종합저층특정,연후장기작위훈련수거,대도상유별진행반감독학습,건립도상화유별적어의영사,최후분별채용이차식거리화개진적 Canberra 거리대도상저층특정진행도량,특정공간중도상류적구역중심용정반궤진행질대경신。통과실험대비,해도상검색산법구유교고적준학솔,우우전통적기우내용적도상검색산법。
A method of image retrieval based on the feature fusion of region of interest was proposed to realize the seman-tic correlation of images content.First, the regions of interest were divided and the integrated underlying characteristics of image were extracted.Second, the characteristics were used as training data to classify the images by semi-supervised learning, then the mapping between images and categories of semantic was established.Finally, the quadratic distance and the improved Canberra distance were respectively used for measuring low-level features, and the cluster centers of images in the feature space were updated iteratively through positive feedback.The experiments compared with other algorithms showed that the proposed image retrieval algorithm had higher accuracy and performed more effectively than traditional algo-rithms.