An image processing method and device combining idle resources of vehicles and machines

Abstract

The present invention provides an image processing method and device combining the idle resources of the vehicle and the machine. The method includes: regularly checking the use state parameters of each hardware device in the vehicle and machine, and obtaining the idleness of the vehicle and machine based on the use state parameters. Resource parameters; judging whether the idle resource parameters meet the preset conditions of the car-machine running a specified scene detection algorithm; if the idle resource parameters meet the preset conditions, run the scene detection in the car-machine The algorithm assists the preset image detection unit to visually detect the image data input to the image detection unit in real time. Based on the solution provided by the present invention, interconnection between the vehicle and the preset image detection unit is provided to provide auxiliary image processing functions, and to provide assistance to the relevant detection algorithm on the preset image detection unit, so as to improve the detection accuracy. At the same time save costs.

Description

technical field [0001] The invention relates to the technical field of automobiles, in particular to an image processing method and device combining idle resources of a vehicle and a machine. Background technique [0002] At present, in the fields of automatic driving and assisted driving, for automatic parking or active safety functions, target detection algorithms are often used to assist in identifying surrounding environment information, while general target detection algorithms are generally implemented through machine learning or deep learning. . For assisted driving functions, vision-based detection algorithms generally run on proprietary chips, but the computing power of proprietary chips is very limited due to cost issues, so the accuracy and speed of vision algorithms often need to be balanced according to actual scenarios , sacrificing a small amount of precision to make the speed meet the real-time requirements, so that it is impossible to effectively balance th...

AI Technical Summary

Problems solved by technology

For assisted driving functions, vision-based detection algorithms generally run on proprietary chips, but the computing power of proprietary chips is very limited due to cost issues, so the accuracy and speed of vision algorithms often need to be balanced according to actual scenarios , sacrificing a small amount of precision to make the speed meet the real-time requirements, so that it is impossible to effectively balance the processing accuracy and real-time requirements of the visual algorithm

Images