CT Image Processing And Parallel Spatial Statistics Of Coal And Rock Mass Based On Hadoop

CT image is made up of a number of different gray scales from black to white pixels according to the matrix form.These pixels reflect the X ray absorption coefficient of the corresponding voxels.CT image can obtain its internal information without destroying the object,this makes the CT scan imaging become an indispensable research method in medicine,geology and other subjects.Currently,Hadoop has become the industry recognized distributed computing platform which can effectively deal with large data,the advantages of Hadoop include high reliability and high scalabilityHadoop has the characteristics of high efficiency of distribution,and it can be used to solve the problems of slow speed and high memory requirement in CT image processing.In order to achieve the above research,this paper makes the following exploration.Useing Hadoop platform to study several methods of improving the recognition of CT images,and uses the computational advantage of Hadoop distributed computing platform to study more efficient parallel spatial statistics.This paper minutelyintroduces the Hadoop platform and its key components,HDFS and MapReduce,also,the working process and methods of HDFS and MapReduce in data processing are described in detail.In addition,this article also introduces the methods of constructing the Hadoop cluster,studiesseveral image enhancement methods for improving CT image recognition,the original data adopts two kinds of image denoising and image enhancement methods which are more suitable for this type of CT image.Finally,the information such as the fracture of coal or rock,rock of 1208 CT images was carried out in parallel spatial statistics,and then we can get more detailed quantitative information from CT images.