In this paper, a multi-instance learning-based CBIR (content-based image retrieval) approach is presented, and multi-instance learning is applied in CBIR in order to deal with the inherent ambiguity of images. First of all, the whole image is regarded as a multi-instance bag. Secondly, the image is partitioned into a number of regions using Adaptive k-means image segmentation method. Then query images posed by the user are transformed into corresponding positive and negative bags and a EM-DD algorithm is employed for image retrieval and relevance feedback. Finally, the users can get satisfactory results.