Extraction of color entropy sequence for micro-vessel detection in virtual slide

摘要

In this paper we proposed a method to extract the entropy sequence of both red and brown colors for micro-vessel detection in the virtual slide. The virtual slides are taken from the double-stained liver tissue which contains several parts of micro-vessel. Our contribution in this paper lies in three aspects. First, both the red and brown pixels in the virtual slide, which indicate the key components of micro-vessel, are represented in some simple rules. Second, the entropy sequences of both red and brown pixels are extracted to highlight the difference between the positive and negative micro-vessel samples. Third, the feature sets on entropy sequences are utilized to identify the virtual slides with micro-vessels using FCM (fuzzy C-means) method. The experimental results show that the performance of the proposed method is promising.