A Batch Recognition and Extraction Method of Floating Objects on Inland Water Surface Based on Vector-Raster Combination

Abstract

The invention discloses a method for batch identification and extraction of floating objects on inland water surfaces combined with vector rasters, which comprises the steps of: acquiring remote sensing images within the water surface range; classifying the remote sensing images within the water surface range, screening out the spots suspected of floating objects on the water surface and exporting , to obtain the surface layer of suspected floating objects; calculate the area of ​​each suspected floating object in the surface layer of suspected floating objects, and obtain the point layer of suspected floating objects; calculate the point layer of suspected floating objects The eigenvalues ​​of the non-water surface floating objects at each point; according to the eigenvalues ​​of the non-water surface floating objects, the misjudgment spots are eliminated, and the identification results of the water surface floating objects are extracted. Its remarkable effect is: making full use of the vector data spatial analysis technology and remote sensing raster data, it realizes the fast and accurate automatic identification of floating objects on the land and water surface in a large scale, and makes up for the foreign object identification method that only uses remote sensing images to identify floating objects. The defects of the same spectrum make the recognition accuracy higher.

Description

technical field [0001] The invention relates to the technical field of geographic information, in particular to a method for batch recognition and extraction of floating objects on inland water surfaces combined with vector and raster. Background technique [0002] With the rapid development of my country's social economy, it has also brought about the intensification of water pollution, and a large number of floating objects have appeared on the water surfaces of inland lakes, reservoirs, and rivers. The existence of floating debris not only pollutes the water environment, threatens human health, but also adversely affects the development of regional economy. [0003] At present, the recognition methods for floating objects on the water surface mainly include background subtraction method and image segmentation method. The extracted water surface area image is subjected to relief processing and edge detection to obtain the position of the target floating object; the mean-s...

AI Technical Summary

Problems solved by technology

The current methods are basically based on image algorithm recognition, and it is difficult to avoid the phenomenon of "different objects with the same spectrum" misjudgment

Images