White blood cell segmentation and extraction based on histogram multi-threshold and lifting wavelet

Cell image segmentation is an important study direction, and it is the base of much after disposal. In order to solve problems of the traditional method for cell segmentation, the lift item of lifting wavelet can be seen as an adjustable parameter. By training, the lift item would own the character of target. Then we can make use of the lifted filter to remove noise and make double-threshold segmentation on the reconstructed picture. The target would be extracted. The experiment results show that the lifting wavelet algorithm can detect the objection well and it provides a speed realization compared with the traditional template matching method.