Obstacle identification method and system

Abstract

The invention discloses an obstacle identification method. The method comprises the following steps: acquiring a first depth image of a surrounding environment in a driving route; performing cutting processing on the first depth image to obtain a second depth image corresponding to the driving lane; performing ground fitting based on the second depth image, determining a ground straight line, removing a ground part according to the ground straight line, and obtaining a third depth image without the ground part; and performing clustering analysis on the third depth image, obtaining a pluralityof obstacle clustering point sets, and determining position information of each obstacle. Correspondingly, the invention further discloses an obstacle identification system. According to the invention, the obstacle is monitored in real time, and the obstacle can be better avoided.

Description

technical field

[0001] The invention relates to the technical field of computer vision, in particular to an obstacle recognition method and system. Background technique

[0002] With the development of science and technology, robots or express unmanned vehicles are more and more widely used. Obstacle recognition and avoidance has become an important embodiment of robot intelligence. With the continuous development of various robots, the requirements for obstacle avoidance are also increasing. The real environment is often complex and changes in real time. It is necessary to be able to accurately identify obstacles and obtain the distance between them and the robot. The patent application number is 201510891318, which is based on the Kinect sensor depth map and the uncertainty map construction method of the robot's working environment. It adopts the technical solution of pre-stored ground for ground detection. When encountering slopes, etc., the ground cannot be detected no...

Images