Abnormal submerging area detection method based on remote sensing vegetation index time sequence

Abstract

The invention discloses an abnormal submerging area detection method based on a remote sensing vegetation index time sequence. According to the technical scheme, abnormal (irregular) submerging areas are directly detected from a vegetation index time sequence image, and submerging area extraction is not needed. The abnormal submerging area detection method includes the steps that firstly, time sequence data are divided into two sections including a historical time sequence and a monitoring period time sequence, wherein a to-be-detected abnormal submerging time period is included in a monitoring period; secondly, a sequence section which is stable and free of abnormal changes is found at the tail of the historical time sequence to serve as a stable historical time sequence, and fitting modeling is carried out on the stable historical time sequence; thirdly, a stable historical time sequence fitting model is used for conducting data forecasting on the monitoring period, a monitoring period time sequence forecasting value is compared with a monitoring period time sequence observation value, and the abnormal change level is accordingly calculated; finally, under the constraints of a certain vegetation index threshold value, the abnormal change level is compared with a preset abnormal change level threshold value, whether abnormal submerging occurs in the monitoring period or not is judged, and the abnormal submerging areas are finally detected.

Description

technical field [0001] The invention belongs to the field of remote sensing image processing and information extraction, relates to remote sensing image change detection and flood submerged area extraction, and in particular to a method for directly detecting abnormal submerged areas based on remote sensing time series images. Background technique [0002] The spatiotemporal dynamics of flooding are critical to the dispersal and distribution of biological populations and nutrients. Under normal circumstances, such as seasonal floods in rivers and lakes, floods spread and recede in a relatively small space. But when there is a major flood, the floodwaters inundate large areas and can have disastrous effects on land cover and human activities. According to whether an area has been inundated regularly in the same period of time in the past few years, inundated areas can be divided into two categories: regular inundated areas and irregular inundated areas. Among them, regular ...

AI Technical Summary

Problems solved by technology

This method avoids the problems of unstable water body identification accuracy and difficult representation of flood dynamic changes in the existing technical methods, and is applicable to the detection of large-scale abnormal submerged areas

Images