Study of colony image segmentation algorithm based on wavelet transform and watershed transform
一种基于小波变换与分水岭变换的菌落图像分割算法研究

The colonies are different in size and shape and disordered distributing. A new segmentation algorithm based on the wavelet transform and watershed transform is presented. Firstly, wavelet transform was used to create multi-resolution images. Then watershed segmentation algorithm was applied to segment the lowest-resolution image and get the initial watershed segmentation. Finally, the low resolution segmentation was projected to high-resolution to get the full-resolution image. Experiments results show that the colony images can be well segmented by the new algorithm.