Marine obstacle recognition method and system based on binocular vision and radar

Abstract

The invention provides a marine obstacle recognition method and system based on binocular vision and radar. The method comprises the steps of obtaining a target object classification label according to a trained deep learning network model and a binocular vision image after time fusion, and carrying out three-dimensional reconstruction on the target object to obtain target object information of the target object under a binocular vision coordinate system; obtaining target object information under a radar coordinate system according to the radar data after time fusion; respectively carrying outspace fusion on the target object information under the binocular vision coordinate system and the radar coordinate system, and carrying out data matching on the target object information after spacefusion to obtain an optimal matching pair of the target object; and marking a classification label on the target object detected by the radar according to the optimal matching pair to obtain a finaltarget object recognition result. The defects of a single sensor are overcome, and the method is more suitable for being implemented in a severe marine environment on the premise that the positioningprecision and the recognition accuracy are ensured.

Description

technical field [0001] The present disclosure relates to the technical field of marine obstacle recognition, in particular to a method and system for recognizing marine obstacles based on binocular vision and radar. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. [0003] Unmanned boats can be used for maritime surveying and mapping, disaster relief, investigation, monitoring, etc., reducing the risk of manual tasks, and have important applications in the civilian and military fields. Autonomous navigation in the case of over-the-horizon and remote control cannot function requires the ability to perceive the environment. Therefore, environmental perception technology is the first link and key technology to realize the autonomous navigation of unmanned boats. Compared with the terrestrial environment, the sea surface has many difficulties such as...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the deficiencies in the prior art, the present disclosure provides a method and system for identifying obstacles at sea based on binocular vision and radar, which provides category information through binocular cameras and makes up for the shortcomings of millimeter-wave radars that cannot detect target categories. Wave radar has high ranging accuracy, which makes up for the shortcomings of low ranging accuracy of visual sensors, overcomes the shortcomings of a single sensor, and is more suitable for implementation in harsh maritime environments on the premise of ensuring positioning accuracy and recognition accuracy.

Images