A Method of Image Fusion Based on Parallel Computing Algorithm

Abstract

An image fusion method based on a parallel computing algorithm belongs to the technical field of image processing. Through the data resampling method of the image data of the two scenes, a two-dimensional data table is used to describe the spatial position and pixel information of the image, and the topological corresponding relationship of the two scene images at the pixel level is constructed through the two-dimensional data table; On the basis of constructing the one-to-one correspondence topological relationship of the image element level of the two scenes, the image data is grouped and sliced ​​according to the row direction (or column direction, or data block), and according to the specific conditions of computing resources, through multi-threading and multi-tasking The method synchronously starts the corresponding calculation processing flow for parallel calculation, and realizes the fast fusion calculation of the two scene images.

Description

technical field [0001] The invention relates to an image fusion method based on a parallel computing algorithm, belonging to the technical field of image processing. Background technique [0002] Since the image fusion technology can not only preserve the spatial details of the high-spatial-resolution panchromatic image, but also retain the characteristic information of the ground object spectrum characteristic of the lower-spatial-resolution multispectral image, it can greatly improve the utilization of image information. Enhancing the reliability of image interpretation and improving the effect of image interpretation are widely used and play a great role in forest resource monitoring, wetland resource monitoring, sandy land monitoring and change analysis. Image fusion technology has developed into a very common basic technology in practical application of remote sensing technology. [0003] However, traditional image fusion algorithms are all applications based on serial...

AI Technical Summary

Problems solved by technology

[0003] However, traditional image fusion algorithms are all applications based on serial computing algorithms based on traditional CPU architecture. For remote sensing image data, with the improvement of spatial resolution, the data volume of each image is getting larger and larger. When performing image fusion, limited by computer resources and traditional serial computing algorithms, it is increasingly unable to meet people's requirements for image data processing speed.

Images