The Segmentation Of Colored Blood Micrograph Based On Region And Edge Information

The count and recognition of white blood cells plays an important role in modern clinical practice. At the same time, it is the key foundation for diagnosing kinds of blood diseases and other pertinent diseases. Those blood corpuscle analyzing apparatuses used currently in hospital have many disadvantages, for example it can't observe the shape of the cells, can't save the swatch of the test, and the machine is very expensive, and so on. So some secondary hospitals still do this work by people. By this token, extracting the leukocytes from the blood cell image and recognizing it using the method of digital image processing and pattern recognition is the leading developing direction in this field. And the auto-recognizing technology for the blood cell image is one of the representative projects.In the auto-recognizing system for blood cell images, the most important step is the segmentation the image and extracting the leukocytes from the background images, and the result directly influence the later processing, such as feature extraction. But at present, we have no segmentation method can be widely taken into practice. So to study an exact and simple algorithm for extracting the leukocytes is a very significant project.The main task of this paper is to study an algorithm for segmenting the blood cell micrograph, and extracting the leukocytes from the background images. The method of wavelet, region growing, color grads, and some method of the binary morphologic image processing are applied in this algorithm system. The algorithm also integrates the color information, the region distributing information, and the edge information to gain its ends. First, divide the whole blood cell image into three regions by multi-scale threshold algorithm. In this step, we get the nucleolus of...