According to target searching model and relevant theories of modern image processing, global similarity coefficients and local similarity coefficients based on gray histogram, gray co-occurrence matrix and gray-grads co-occurrence matrix were brought forward as two important parameters to evaluate stealth effect of infrared coating.

Correlativity between summer (June -August) precipitation and earlier potential height field at 500hpa of Northern Hemisphere was analyzed by using the local pattern analogue coefficient (LPAC) and the differance chart, Key area shown in LPAC chart were attempted for precipitation forecast in summer and better effect was gotten.

And a method for very short-term load forecasting is presented based on the local shape similarity.

A recommendation algorithm based on users partial similarity;

This paper proposes a method for sub-pixel edge location based on the partial similarity according to the partial similarity of the pixels gray feature,analyzes the distribution change characteristic of LSM through the sub-pixel edge location model,detects the edges with amplitude and extracts the sub-pixel edge position with conic fitting.

The local correlation coefficient is introduced and the different thresholds at different levels of wavelet decomposition are set up, in order to reduce the spectral distortions of the fusion image,fuse multi-spectral images with a panchromatic image of Quickbird data by the proposed method.