Study Of Flow Characteristics Extraction Technology In Microcirculation Based On Image Processing

Microcirculation, a main material exchange place for organism, plays an important role in biological life activities. Studying the flow of blood in microcirculation, and extracting and analyzing the blood flow characteristics in micro vascular have great application prospects in revealing certain physiological mechanism, testing and analyzing illnesses, etc. And the study of flow characteristics is one of the key issues in medical field, computer vision field and image processing field. At present, with the great acceleration of computer processing speed, the method of combining machine vision with digital image processing technology is applied to the field of micro vascular cell detection and tracking, which not only causes no wound to the organism, but also accelerates date processing speed. Therefore, the use of machine vision and image processing technology to complete the function of obtaining blood flow characteristics in microcirculation has great significance in the biological sciences, medical diagnosis and so on.Tracking cell movement in micro vascular is important to get various parameters of blood flow characteristics. However, it is really hard to track the movements of red blood cells automatically in image sequences. The main reasons are erythrocyte morphology variability and position changes of vascular due to breathing movement and other external factors. Because of the collision among erythrocyte, leucocyte, thrombocyte and other particles, blood often flows in the form of inner cell mass. In view of this, a new method based on image sequences to achieve biological characteristic acquisition of microcirculation is proposed in this paper. The microscopic images of circulating blood cells are obtained by combining electron microscopy with machine vision. The obtained digital image information is processed by using digital image processing technology. Inner cell mass is extracted by combining grayscale histogram with threshold. Inner cell mass is tracked by template matching method in image sequences of blood flow. And then, the centroid coordinates of inner cell mass is calculated by the centroid algorithms. Finally, the result of the blood flow velocity can be worked out. The operations of image format conversion, threshold segmentation, tracking, date extraction can be accomplished by software programming. The method proposed in this paper not only extracts the blood flow parameters, but also provides a new research method in the biomedical domain.