Multi-scale change monitoring of water environment using cloud computing in optimal resolution remote sensing images

Abstract

Water is the source of life and a very important part of life. However, the current water resources security is facing global challenges. Affected by climate change and human factors, how to quickly realize the continuous monitoring of water resources changes is crucial to the sustainable management of water resources. In recent years, the amount of multi-temporal remote sensing data has grown rapidly, which has caused problems such as weak single-computer processing for change detection and difficulty in mass image storage management. Traditional single-computer processing methods can no longer meet the needs of remote sensing big data change detection. It provides cutting-edge methods and means for monitoring large-scale changes in water resources. Therefore, a distributed parallel method for remote sensing image change detection based on cloud computing is proposed. Through this method, the change detection efficiency of massive remote sensing images is improved, and the high-resolution remote sensing image change detection based on fully connected conditional random fields is deeply analyzed. By summarizing the experimental data of this method and comparing with other data, it can be seen that when the parallelism is higher, the advantage of running time is more obvious. The performance improvement is also as high as 89.49%. Therefore, distributed parallel processing through cloud computing can complete the efficient change detection of massive remote sensing images without affecting the accuracy. This helps us to better monitor and monitor water resources at multiple scales.