Map/Reduce in CBIR application

Since the birth of the concept of Content Based Image Retrieval, it has been a hot issue of computer vision and pattern recognition subject. In the information age, with the rapid growth of the data, how can we retrieval image quickly and accurately from the vast amounts of image data become a problem. This paper proposes a method of applying the Map/Reduce model in the Cloud Computing to Content Based Image Retrieval. Through experiments on the image library of different sizes, we confirmed that this method can greatly improve the efficiency of image retrieval on massive image data.