An underwater fishing net detection method based on forward-looking sonar images and auv platform

Abstract

The invention discloses an underwater fishing net detection method based on a forward-looking sonar image and an AUV platform. The multi-beam forward-looking sonar mounted on the AUV collects and obtains the original data packet of the forward-looking sonar; the obtained sonar original data packet Perform real-time analysis to obtain the original sonar data matrix, and then interpolate the sonar data to obtain a complete fan-shaped forward-looking sonar image; transmit the obtained forward-looking sonar image to the AUV platform, and input it to the pre-trained target detection Fishing net detection is performed in the model in real time; the detection results are converted into corresponding system instructions, and fed back to the AUV main control module, and the AUV makes corresponding obstacle avoidance decisions based on the detection results. This solution solves the problems of complex algorithm implementation, low detection accuracy, and poor real-time performance in the existing technology, improves the autonomous survivability of the AUV, enables the AUV to grasp the information of the sea area ahead in real time, and can independently detect and monitor underwater fishing nets. dodge.

Description

technical field [0001] The invention relates to an underwater fishing net detection method based on a forward-looking sonar image and an AUV platform, and belongs to the technical field of forward-looking sonar image target detection. Background technique [0002] Autonomous underwater vehicles (AUVs) are important tools for humans to explore the ocean. It can be widely used in marine survey, resource exploration and military fields. Inspired by the human visual system, the vision system of underwater robots mainly relies on sonar, and forward-looking sonar (FLS) is one of the main sensors for AUVs to detect underwater targets. The scene detected by forward looking sonar can be visualized as a forward looking sonar image. In this way, AUVs can identify and detect targets from sonar images and perform various marine tasks, such as path planning, underwater archaeology, fish identification, etc. Therefore, we can let AUVs utilize FLS images to detect and locate underwater F...

AI Technical Summary

Problems solved by technology

[0010] Aiming at the deficiencies of existing target detection algorithms, the present invention proposes an efficient and accurate target detection algorithm suitable for forward-looking sonar images. The algorithm can be embedded in the AUV system to overcome the real-time detection and autonomous survivability of existing AUVs. In the complex underwater environment, it can effectively improve the detection efficiency, achieve real-time performance, and also have high detection accuracy, so as to realize the real-time detection of underwater fishing nets by AUV in complex sea areas, and provide guarantee for AUV obstacle avoidance decision-making

Images