Hyperspectral image river detection method

Abstract

The invention relates to the technical field of remote sensing image processing, in particular to a hyperspectral image river detection method, computer equipment and a computer readable storage medium. The hyperspectral image river detection method includes the steps: inputting a hyperspectral image; filtering the hyperspectral image; decomposing the image, and calculating shear wave characteristics under different scales and different directions; normalizing all the shear wave characteristics, and calculating the standard deviation of each shear wave characteristic; selecting two shear wavefeatures with the maximum standard deviation under each scale for summation, and obtaining a feature image of the river; performing adaptive threshold segmentation on the feature image to obtain a binary image; and taking the binary image as an active contour model template, and performing river detection by using the active contour model to obtain a river detection result. According to the hyperspectral image river detection method, the use of edge information and learning methods is avoided, and the problem of difficult river target detection under a small sample is solved, and the detectioncapability of a river false alarm source can be significantly improved, and the calculation efficiency is improved at the same time.

Description

technical field [0001] The invention relates to the technical field of remote sensing image processing, in particular to a hyperspectral image river detection method, computer equipment and a computer-readable storage medium. Background technique [0002] Hyperspectral images have a wide range of applications, such as security monitoring, geographic information system updates, environmental detection, etc. Therefore, the extraction of different types of targets in hyperspectral images has become a research hotspot in recent years, and the extraction of rivers is one of them. question. Due to the long imaging distance and low imaging quality of hyperspectral images, the targets in the images are easily confused with other interference information such as the background, which makes detection more difficult. [0003] At present, river detection methods mainly use river edge information or complex mathematical models such as snake or genetic models to extract rivers, which req...

AI Technical Summary

Problems solved by technology

Due to the long imaging distance and low imaging quality of hyperspectral images, the targets in the images are easily confused with other interference information such as the background, making detection more difficult

[0003] At present, river detection methods mainly use river edge information or complex mathematical models such as snake or genetic models to extract rivers, which requires a large amount of calculation and low efficiency.

In addition, remote sensing image target detection often uses fusion methods, but most of the fusion methods are to fuse different types of raw data, or to fuse a large number of processing results at the decision-making level, and the number of hyperspectral images is relatively large. less and poor imaging quality, it is difficult to obtain more accurate detection results through existing fusion methods

Images