Target detection method and device for remote sensing image, equipment and storage medium

Abstract

The embodiment of the invention provides a target detection method for a remote sensing image, and the method comprises the steps: obtaining a to-be-detected remote sensing image, inputting the to-be-detected remote sensing image into a pre-trained target detection model, and carrying out the target detection, and obtaining the position information of a first detection frame in the to-be-detected remote sensing image, wherein the confidence coefficient of the preset target in the first detection frame is higher than the first confidence coefficient; intercepting candidate images from the remote sensing image to be detected according to the position information of each first detection frame; inputting the candidate image into a target detection model for target detection, obtaining position information of a second detection frame in the candidate image and taking as a target detection result, wherein the confidence coefficient of a preset target in the second detection frame is higher than a second confidence coefficient, and the second confidence coefficient is higher than the first confidence coefficient. Thus, when target detection is carried out for the second time, the resolution of the candidate image is low, a good target detection effect can be obtained, and therefore high-precision target detection is carried out on the remote sensing image.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a target detection method, device, equipment and storage medium for remote sensing images. Background technique [0002] Typhoons are one of the most important meteorological disasters. The disasters caused by them will cause serious damage to people's lives and economic development. Through the forecast and early warning of typhoon weather, people can deploy and avoid them in advance to reduce the losses caused by typhoons. [0003] In the prior art, meteorological remote sensing satellites can be used to observe cloud cover and ground surface features on the earth from top to bottom, and obtain satellite cloud images, in which typhoon cloud clusters can be observed. In this way, staff can use Davorak and DAV -T (Deviation Angle Variance Technique, deviation angle variance) technology, compares the typhoon cloud cluster in the satellite cloud image with the standard typ...

AI Technical Summary

Problems solved by technology

[0004] However, in the above method, it is usually necessary to manually locate the typhoon cloud cluster in the satellite cloud image, and the cost is relatively high

However, when using the target detection model in the prior art to locate typhoon cloud clusters in satellite cloud images, due to the high resolution of satellite cloud images and the large target detection range, it is easy to miss detection, and the satellite cloud images also include a large number of other non- Typhoon cloud clusters have a lot of noise information, which can easily lead to inaccurate target detection results

Images