Image annotation,a technique for connecting image semantics and visual features,can be used to present image semantics well.A method for image annotation using relevance feedback log and semantic network is presented.Firstly,the image semantics are acquired by users'relevance feedback log;then semantic clustering is carried out based on semantic similarity;lastly automatic image annotation is realized through semantic propagation.The experimental results indicate that increasing images will be annotated as increasing relevance feedback log and the annotation accuracy will be stable with the increment of relevance feedback.

Establishing a mapping relationship between image visual features and semantics can be used to reduce gap. A novel mapping method for image semantics and visual features was presented. In this method, image semantic information could be captured by adding users' relevance feedback, and then a decision table of visual features and semantics was constructed. Knowledge reduction of rough set theory was used to reduce the redundant visual features according to semantics, by doing that a mapping relationship between image visual features and its semantics was established. The experimental results indicate that amount of visual features irrelevant to image semantics can be reduced greatly, the complexity and cost of semantic classification are reduced and the accuracy of classification is better.

In order to overcome the defect that the conventional color histogram retrieval method is prone to lose the spatial information of colors,an image retrieval method based on the largest color-connected regions was presented,and its rim roughness to capture the regional traits and reduce the absolute dependence on color.Then the similarity of the images is defined as the combination of the similarity of an integer and the similarity of the blocks to reflect the recognition process by people.The experimental result on real-world image collections indicates that the proposed approach significantly improves image retrieval accuracy.

Introduces basic principle of lifting scheme and presents method of construction of traditional wavelets via lifting scheme.An adaptive threshold based on lifting wavelet transform for image denoising is studied.This method is derived in a Bayesian framework and threshold is chosen according to different subbands and orientations.Comparing with traditional denoising methods,this method combined with soft threshold algorithm,can improve the PSNR more effectively and also makes denoised image more clearly.Adaptive threshold based on lifting wavelet transform,which can be computed fast with a simple implementation,has a good effect for image denoising.