Vehicle-mounted detection optimizing method based on combination of images of adjacent frames and single frame of image

Abstract

The invention provides a vehicle-mounted detection optimizing method based on the combination of images of adjacent frames and a single frame of image. The experiment result shows that the detecting speed of the single frame of image is greatly increased, and the images of the adjacent frames are optimized. On the premise that the calculated amount is reduced, the performance is very close to the performance obtained by the detection on the single frame of image based on a statistics learning method, and time is relatively shortened by near three times.

Description

technical field [0001] The invention belongs to the technical field of video image processing and pattern recognition, and in particular relates to an optimization method for pedestrian detection based on vehicle-mounted multi-frame images. Background technique [0002] With the rapid development of the national economy and the rapid increase of motor vehicles, my country's urban traffic problems are becoming increasingly serious. In a mixed traffic system environment, pedestrians and vehicles are the main components of traffic. As an important factor affecting social stability, traffic safety has been paid more and more attention by the society. If we apply pedestrian detection and tracking to the assisted driving system of smart cars, detect pedestrians in front of the vehicle, promptly inform the driver that there are pedestrians in front of the vehicle, and provide safety warnings, which will help reduce or avoid collisions between vehicles and pedestrians. great signi...

AI Technical Summary

Problems solved by technology

But its disadvantage is that the complexity of the algorithm is high, and it takes a lot of time to detect and identify pedestrians in a single frame image, so that it is not suitable for real-time detection of multiple frames.

Images