Research On Retinal Cell Count Based On Adaptive Filtering And Intelligent Algorithm

As the basic unit of the life body,the function of the cell is self-evident.So cell technology is born,Cell technology is based on microscopic images of cells.Therefore,microscopic image processing and analysis of cells has become an indispensable technology in the field of life sciences,Occupy a pivotal position.With the development of science and technology and the maturity of microtechnology,the microcosmic data we have obtained are becoming more and more abundant.However,the accurate study and analysis of these data is a challenging task for people in the field of life science and the field of computer science.A number of expert studies have found that the number of cells in the area can be calculated to know whether the part of the body is healthy or not.So cell count has also become an important way to study cell images.The aim of this paper is to detect and count the cells in the field of microscopic image processing.The target is the high-density cell in the cat retina under the dark field.The main contents of the paper are as follows:Aiming at the image characteristics of high density retinal cells(dark areas in the cell and high brightness around them),we first conduct one filtering in the frequency domain,Two-dimensional discrete Fourier gaussian high-pass filter,and enhance and sharpen the image.Secondly,we have two new filters--threshold division adaptive fuzzy enhancement filter,threshold division adaptive suppression gradient filtering,they are used to turn the cell center dark area into bright area,while reducing the rest of the brightness.In order to filter the cells effectively.In the aspect of cell segmentation,this paper is based on the maximum value processing based on intelligent algorithm.In this paper,the random adaptive extremum calculation is added to improve the computational efficiency and optimize the processing results.In addition,we take two improved intelligent algorithms--genetic algorithm and particle swarm optimization.We combine the improved extreme intelligence algorithm with the intelligent algorithm to segment and count the image.Finally,by comparing with artificial counting and error analysis,it is concluded that the error rate of this processing result is low,and it is completely in the acceptable range of error.It is proved that the method proposed in this paper is suitable for the image processing of high-density retinal cells.