Ground identifier category confirmation method and device and vehicle

Abstract

The invention provides a category confirmation method and device for ground identifiers and a vehicle, and relates to the technical field of computers. The method comprises the following steps: acquiring an image comprising a to-be-determined ground identifier; performing inverse perspective transformation on the image to obtain a top view corresponding to the image; and determining the category of the to-be-determined ground identifier by using the top views and a ground identifier identification model, training the ground identifier identification model by using a top view set, and obtaining each top view in the top view set by performing inverse perspective transformation on an image comprising a ground identifier with a known category. By using the method provided by the invention, the detection capability of the ground identifier is improved, and especially the detection capability of the far-end ground identifier is improved, so that better technical support is provided for path planning, real-time decision making and driving control in the subsequent automatic driving process, and the method has relatively high practicability.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to a method, device and vehicle for confirming the type of ground signs. Background technique [0002] In autonomous driving technology, environmental perception is the first link, and its core is to enable the vehicle to better simulate the perception of human drivers and accurately understand the surrounding driving conditions. In practical applications, accurately realizing the detection of the ground signs of the vehicle driving road is of great significance for subsequent path planning, real-time decision-making and driving control. [0003] At present, the method for identifying the category of ground markers is generally a method based on the Region of Interest (ROI) and artificially constructed features, such as the method for confirming the category of the ground marker horizontal projection. The method uses the histogram of oriented gradients (HOG) to extr...

AI Technical Summary

Problems solved by technology

[0004] However, this method is only suitable for dealing with the situation when the ground mark is located at the near end of the current vehicle. This is because the position of the camera device of the vehicle is fixed and is affected by the shooting angle of the camera device. For the ground mark at the far end of the vehicle, the horizontal projection detection The obtained candidate ground markers generally have serious deformation and distortion, which makes it difficult to accurately determine the type of ground markers

Therefore, the ground mark detection ability of this method is poor, and the ground mark detection can only be realized in the vicinity, and the practicability is low

Images