The Study Of Content-based Diversity Plant Image Retrieval

In the previous studies, only the relevance is cared in retrieval system, so manyduplicate or near duplicate documents retrieved in response to a query. To solve thisproblem, the diversity retrieval technique has proposed. We do deep research onclustering diversity retrieval, feature extraction method on multi-manifolds learning.The main work for this paper can be summarized as follows:(1) Verify the effectiveness of a variety of image features in content-basedplants image retrieval. Through the characteristic of different image, we extractdifferent feature in diversity retrieval. Such as extract shape and color feature inflower image, extract shape and texture feature in leaf image.(2) Proposed plant image diversify retrieval based on maximal scatterevaluation. We use the relevance feedback technique based of SVM, and then usethe Affinity Propagation clustering algorithm to make the retrieval result not onlyrelevance, but also diversity. We also proposed a new evaluation function-MaximalScatter Diversity (MSD) static evaluation function.(3) Proposed diversity image retrieval by feature extraction on Multi-manifoldsLearning. In high dimensional space, each class data represents a manifold, so themany class type of image feature cast to low dimensional space should keepdifferent manifold far away, and expand the distribution between different submanifold, it is conducive to the diversity retrieval. It does not require relevancefeedback technique, also can keep the retrieval result diversity, and also improve theresult relevance.(4) Develop the content-based diversity plant image retrieval system. Thesystem implements a variety of image feature extraction, also achieves the diversityretrieval based on clustering technique and Multi-manifolds learning.