Multi-view road intelligent identification method based on dual-light fusion

Abstract

The invention discloses a multi-view road intelligent identification method based on dual-light fusion. The method comprises the following steps: collecting a visible light image and an infrared image of a road; preprocessing and fusing the visible light image and the infrared image to obtain a fused image; performing three-dimensional reconstruction on the visible light image and the infrared image to obtain a three-dimensional point cloud image; and training and detecting a pre-constructed deep learning network according to the fused image and the three-dimensional point cloud image to obtain a road recognition result; forming a depth map through image fusion and multi-view synchronous photography so as to detect the accurate size, position and distance of a target. When the system is applied to an automatic driving scene, the vehicle can detect any object under any road, light and weather conditions, and the all-weather application scene of the automatic driving system is greatly expanded.

Description

technical field [0001] The invention relates to stereo vision technology, in particular to a multi-eye road intelligent recognition method based on dual light fusion. Background technique [0002] Since stereo vision technology can provide accurate depth information and perform 3D modeling, it has been regarded as a crucial vision solution in the fields of autonomous driving, robot navigation, AGV and other environmental perception. Compared with other vision technologies, stereo vision technology does not use active emission sources, does not need to fuse other sensor data, and has excellent performance in 3D distance measurement, self-position recognition, collision avoidance, object tracking, lane recognition, road boundary recognition, etc. , and the processing speed can also meet the needs of real-time applications. [0003] In order to solve the problem of "blindness" of the low-light stereo vision system under extreme weather and light, infrared imaging technology wi...

AI Technical Summary

Problems solved by technology

Stereo vision systems based on low-light cameras will have the problem of "blindness" in extreme weather and complex lighting. Millimeter-wave ranging and speed-measuring radars cannot finely measure the 3D information of road targets. Emerging lidar stereo vision systems also face the same problem in extreme weather. failure problem

In addition, single-purpose road target recognition relies on deep neural network processing technology, and will encounter extreme situations where unknown objects exist in the training network, especially in extreme weather and complex lighting conditions. The accuracy of target recognition is significantly reduced

Images