Considering that the high-frequency sub-band can gather most energy of noise by the wavelet decomposition, and the
local variances in edge and noise change inconsistently, the real edges of images are separated from the noise by
choosing an appropriate threshold value in each high-frequency sub-bands of wavelet domain, while the WGMAP
method is used to restore the images. Experimental results show that the images reconstructed by the improved algorithm
reproduce preferably the edge structures of original images; and the signal to noise ratio and the visual effect are
A method of locating eyes in a face image is presented in this paper. Compared with most methods in this field, this
method is insensitive to the rotation of image. And when applying this method to the color image, eyes can be
detected in a profile face image although with some false alarms. When dealing with grey-level images, some
geometry features related to eyes are generalized to locate eyes more precisely. When dealing with color images, a
color transforming is utilized to detect the faces and scleras.
Image segmentation is one of the most difficult and important steps in image analysis. Segmentation accuracy determines the eventual success or failure of computerized analysis procedures. The thresholding method is considered as one widely used method in image segmentation. It is considered by Kittle and Illingworth that the optimum threshold of segmentation can be calculated by the method of probability statistics, if the mixture density function composed of object and background is known and estimated. Suppose that the gray histogram of image obeys Poisson distribution, a new method of segmentation based on the maximum relationship principle of conditional distribution is proposed in this paper. The new method is compared with Ostu method and the minimum error method, the result shows that the proposed one is good and the performance is stable.