Research on Image Analysis and Processing Technology Based on Big Data Technology

Abstract

With the increase of information, the traditional image processing technology still has a big shortage in image quality and processing speed. Therefore, how to improve the image quality and processing speed is a research hotspot. To solve the above problems, we adopt the method of combining the generative countermeasure network with the super-resolution reconstruction, and use the open data set to train the model. Through the reasonable selection and optimization of the image processing algorithm, we can achieve the goal of improving the image super-resolution reconstruction effect. In order to improve the slow image processing speed, we set up Spark cluster to integrate computing resources. The experimental results show that the method based on the generation of countermeasure network has a good effect for high resolution image reconstruction, and this method can expand the resolution of low resolution image by four times; The cluster based on the Spark platform shows better performance advantages in processing a large number of image data. When the hardware indicators of the cluster environment and the stand-alone environment are consistent, the cluster based on the Spark platform can increase the running speed of the program by about 10%. When the data volume reaches the tera-byte level or even the petabyte level, the operation speed of the system can still be improved by increasing the overall running memory of the cluster.