Image Segmentation Of Furnace Velocity Field Based On Improved FCM Algorithm

Image segmentation is a key step in image analysis and prcessing,and is widely used in many fields,such as medical treatment,machine vision,rempte sensing image analysis and so on.In this paper,an improved fuzzy C-Means clustering(FCM)image segmentation algorithm based on improved ecekoo search algorithm is prpposed,which is used for image segmentation of furnace velocity field in thermal power plants,and provides a theoretical basis for the application of image processing technoligy in the analysis the aerodynamic field in the furnace,The main work of this research is as follows:(1)In this paper,the euekoo search algorithm is studied and the advantages and disadvantages of this algorithm in solving the optimal value problem are discussed.Aiming at the problem that the discovery probablity in cuckoo seanch algorithm is a fixed value,which affects the optimizing speed of the algorithm,a variable discovery probability formula is defined in this paper,and an improved euekoo search to test the algorithm is proposed,There are several commonly used test funetions selections to test the performance of the improved algorithm.The test results shows that average running speed of the improved algorithm is increased by 5%,and the average objective function value is optimized by 6.93%.(2)Based on the analysis of the advantages and disadvantages od traditional segmentation methods in the image segmentation of velocity field,the superiority of clustering methods in the segmentation of such images is summarized in this paper,and then the application of FCM clustering algorithm in the image segmentation of furnace velocity field is focused on.Aiming at the problem that the FCM algorithm is sensitive to the initial clustering center and easy to fall into local optimum,a FCM image segmentation algorithm improved cuckoo search has satisfactory segmentation quality with the average Dice_ratio coefficient of 89.97%,the average Jaccard coefficient of 86.28%,the Precision of 88.74%,and the Recall rate of 91.76%.(3)The velocity field simulation experiment of four-corner coal-fired boiler is carried out based on FLUENT,and the two-dimensional velocity field simulation image is obtained.The FCM image segmentation algorithm based on cuckoo search is successfully used to extract the center tangent circle area and inlet wind angle area of velocity field image.It is concluded through experiment that the velocity field image is better segmented when the clustering number is 5?7,and it is verified that the improved FCM algorithm is better than the traditional image segmentation algorithm for the segmentation quality of the furnace velocity field image.Based on FLUENT,the velocity field simulation experiment was carried out.The MATLAB platform was used to segment the velocity field simulation image.And the segmentation quality comparison experiment verified that the proposed segmentation algorithm in this research is effective and feasible.