Research On Flame Image Segmentation Of Alumina Rotary Kiln Based On FCM

Accurate judgment of the combustion state in the rotary kiln is one of the important steps in the production of rotary kiln.Correct identification of the combustion condition of the rotary kiln flame can not only increase the production efficiency of the rotary kiln,save costs,improve the production quality of the product,but also guarantee.The personal safety of the staff reduces the amount of pollutants discharged during the production process.With the deepening of research on computer software technology,the use of computer technology to solve the problem of identifying the flame image of rotary kiln is one of the main methods of industrial production.In this paper,the identification of the flame image of the rotary kiln is studied,and the application of image segmentation in the flame image of the rotary kiln is analyzed.The commonly used algorithm is used to segment the flame image of the rotary kiln.Aiming at the poor effect of traditional fuzzy clustering algorithm(FCM)on flame image segmentation,the improved algorithm is applied to the flame image.The effectiveness of the improved algorithm and the accuracy of flame image segmentation are obtained through experimental analysis and comparison.The main research work is as follows:(1)Analyze the working process of the rotary kiln and expound the importance of studying the image segmentation of the rotary kiln flame.A brief analysis of the problems of large lag,multivariable,strong coupling,nonlinearity and flame detection and identification of the rotary kiln,as well as an accurate description of the traditional detection and identification method of the rotary kiln flame image,for the traditional detection.The shortcomings of the method,thus giving the advantages of computer application technology in the image recognition of the rotary kiln flame,lead to the application of the rotary kiln flame image of each algorithm.(2)Using a variety of algorithms to segment the flame image of the alumina rotary kiln.The rotary kiln flame image is more complicated and has a higher degree of pollution.It is used to study the flame image algorithm commonly used in rotary kiln.The definitions and operation steps of each algorithm are given in detail.The advantages and disadvantages of each algorithm are analyzed and explained.Several algorithms are selected to segment the flame image of the rotary kiln,and the results of image segmentation are analyzed experimentally.(3)Apply the energy noise-based FCM improvement algorithm to the alumina rotary kiln flame image.Considering the complexity of the flame image and the environment generated,the traditional fuzzy clustering algorithm only considers the Euclidean distance of the pixel.Ignoring the spatial correlation of the image,the accurate segmentation effect cannot be achieved.On the basis of the traditional clustering algorithm,An FCM segmentation algorithm based on energy detection is proposed.By introducing the energy curve function,the probability of pixel points becoming noise points is designed,and the spatial correlation of pixel points is increased.Secondly,based on the weighted filtering,the spatial information enhancement algorithm of pixels is combined.The noise resistance and the constraint function of the spatial distance are designed.By deducing the new membership function,the effectiveness of the algorithm on the segmentation of the flame image is improved.(4)An improved algorithm combining FCM and MRF based on kernel function is applied to flame splitting of alumina rotary kiln.The kernel function fuzzy clustering algorithm combined with spatial correlation has higher segmentation accuracy when segmenting the rotary kiln flame image,but the operation speed is lower,and the improved FCM is combined with the Markov random field(MRF).The flame image is preprocessed by the spatial FCM algorithm combined with the kernel function,then the FCM and MRF constraint fields are obtained,and finally the maximum probability of the flame image is segmented.The segmentation time and segmentation of the rotary kiln flame image are accurately determined by several algorithms.The degree of comparison has fully verified the efficiency and practicability of the improved algorithm in the rotary kiln flame image.