A New Clustering Algorithm for Noisy Image Retrieval

摘要

The paper concerns an open problem in the area of content based image retrieval (CBIR) and presents an original method for noisy image data sets by applying an artificial immune system model. In this regard, appropriate feature extraction methods in addition to a beneficial similarity criterion contribute to retrieving images from a noisy data set precisely. The results show some improvement and resistance in the noise tolerance of content based image retrieval in a database of various images.