Detection And Recognition Of Flowing Microspheres In Two-dimensional Light-scattering Video

The light scattering technique has many advantages,such as fast detection,high accuracy and good repeatability,especially its label free property.It has been developed rapidly in recent years and widely used in the field of biomedicine.The two-dimensional scattering patterns of the cells or microspheres can be obtained by this method of label free light scattering technique.These two-dimensional light scattering patterns contain abundant physiological and pathological information of biological cells.Through the analysis and processing of the two-dimensional light scattering pattern,we can achieve the cells count,classification of cells and other functions.Flow cytometry can be used to obtain a large number of cells' test results efficiently.How to realize the automatic detection and recognition of two-dimensional light scattering pattern in the flowing video is an important problem to be explored in this paper.In this paper we apply computer vision technology to the detection and recognition of two-dimensional light scattering particles in flowing video.The detection and recognition technique of moving target is one of the important research area in computer vision system.It has been widely used in biological medicine and other fields and it has a great development prospect.By combining the detection and tracking algorithm of moving targets with light scattering technology,we propose a method for analysis of flowing particles based on two-dimensional light scattering video.In this paper,microsphere is used as a model to simulate ideal biological cell and then the two-dimensional light scattering video of the flowing microspheres is captured.We detect and track the flowing microspheres in the video with OPENCV video surveillance system.And then,we extract the two-dimensional light scattering pattern characteristics of each microsphere,which we can be used to achieve the function of the counting as well as the recognition of microspheres.In this paper we introduce the basic knowledge of light scattering first,and explain the occurrence mechanism of the two-dimensional light scattering pattern of particles further;then we focus on the detection of flowing microspheres in two-dimensional light scattering video;on the basis of the realization of the two-dimensional light scattering flowing target tracking,the recognition of the flowing microspheres in the two-dimensional light scattering video is carried out by means of the streak feature extraction algorithm of light scattering pattern.This work lays a foundation for the automatic detection and recognition of complex cells in the future.