Abstract: Lymph nodes are an important organ of the human body immune response．The pathological changes of lymph node are an important basis of malignant tumor detection and judgment of metastasis of cancer (lung cancer,colorectal cancer,breast cancer,liver cancer,cervical cancer,etc.) A segmentation algorithm based on maximal variance between-class and morphology was introduced to segment lymph node．Maximal variance between-class method was used to operate binary enhancement processing for the original image．Mathematical morphology was introduced to do boundary correction for binary image．Erosion and expansion operation was used to solve the problem of the target area connected with excess tissue after binarization．In the end,useful lymph node tissue was better extracted．Experimental results show that the proposed method can get better segmentation effect for lymph node images with surrounding tissue adhesions and larger gray level difference between target and background.

Key words: Maximal variance between-class,Mathematical morphology,Lymph node image segmentation,Enhancement processing