Automatic leukocytes classification by distance transform, moment invariant, morphological features, gray level co-occurrence matrices and SVM

作者：Pang Gai^*; Zhuang Yangkai; Zhou Ping

来源：1st International Conference on Information Sciences, Machinery, Materials and Energy (ICISMME), 2015-04-11 to 2015-04-13.

• Leukocytes classification
• Euclidean distance transform
• Moment invariant
• Morphological features
• Texture features
• Support vector machine

Leukocyte is an important part of the immune system. According to the problem that manual operation is not efficient, a novel automatic classification of leukocytes is proposed in this paper. First, moment invariant based on Euclidean distance transform is extracted from nucleus area and morphological features are extracted from segmented cells. Then, monocytes, lymphocytes and basophils are distinguishing from the other samples using features extracted from the first step. Next, the gray level co-occurrence matrices (GLCM) are used as texture measure to classify the remaining classes via a Support Vector Machine (SVM). Experimental results show that the proposed approach provided good classification accuracy, and sufficiently fast to be used in hematological laboratories.

• 出版日期2015
• 单位浙江理工大学