The Microscopy Serial Image Synthesis Based on the Wavelet Analysis

Due to the limitation of the microscopic focal length, the obtained microscopic images are often blurred in some local areas, which causes that people can not recognize or understand it completely. Based on the different energy distributions of the low-frequency band and the high-frequency band, which are achieved by discrete wavelets transform, an original method for wavelet coefficient combination is proposed in this paper. It takes the ratio of the two bands as a characteristic value and the maximum value is chosen as an integration rule for determining new wavelet coefficients. This method can remove the defocused areas and reduce the distortion of the fused images. Finally, the advantage of the proposed fusion approach is demonstrated clearly by practical experiments, by comparison with conventional area based pixel selection method. From the microscopy full-focus image, researchers can measure the desired results directly.