A neural network model training and universal ground wire detection method

Abstract

The invention discloses a neural network training method and a method for carrying out universal ground wire detection by adopting the neural network, and belongs to the field of intelligent driving.In the prior art, the boundary speed is difficult to estimate; In order to solve the technical problem that the use of a planning algorithm is not facilitated, the neural network and the detection ofa universal grounding wire by using the neural network system provided by the invention comprise the following steps: 1, carrying out single target determination on an image obtained by a camera device, and recording and storing internal parameters and distortion parameters of the camera device; And step 2, inputting the image obtained in the step 1 into the trained neural network to obtain a drivable area segmentation map, a grounding point and a grounding wire. According to the method, the current road image is divided into the drivable area and the obstacle area, the grounding wire and theobject category corresponding to the grounding wire are detected, and detection is more accurate and faster compared with a traditional method.

Description

technical field [0001] The invention belongs to the field of intelligent driving, and more specifically relates to a general ground wire detection method. Background technique [0002] With the development of science and technology, the concept of autonomous driving has been proposed. In the field of automatic driving, the vehicle can be preset with an intelligent system to detect the current drivable area and drive according to this area. [0003] Using the existing drivable area detection method, the image can be processed to obtain the obstacle area and drivable area. However, the prior art cannot identify the type and number of obstacles. This makes it difficult to estimate the velocity of the boundary, which is not conducive to the use of planning algorithms. Contents of the invention [0004] In view of the above defects or improvement needs of the prior art, one aspect of the present invention provides a training method for a neural network model, characterized i...

AI Technical Summary

Problems solved by technology

However, the existing technology cannot identify the type of obstacles and the number of obstacles

This makes it difficult to estimate the velocity of the boundary, which is not conducive to the use of planning algorithms

Images