A Method for Measuring Vehicle Outline Dimensions Based on Image Dynamic Feature Tracking

Abstract

The invention relates to a vehicle outer contour dimension measuring method based on image dynamic feature tracking and a system thereof. The method comprises the steps that step 100) image information is acquired, and images of the vehicle detection process are acquired; step 101) the feature points of adjacent frames of images are acquired based on the images of the vehicle detection process; and step 102) matching of the feature points is performed, splicing is performed on the matched images to acquire the complete image of a vehicle, and the pixel dimension is calculated based on the complete image of the vehicle so that the actual outer contour dimension of the vehicle is acquired. The step 101) concretely includes the steps that step 101-1) the images in which the vehicle head and the vehicle tail are positioned are obtained from the acquired images through segmentation of the foreground and the background of the vehicle; and the vehicle head and the vehicle tail images respectively act as a start frame and an end frame so that the image sequences in measurement of the vehicle are acquired; and step 101-2) traversing of the image frames is performed to acquire candidate feature points, and the feature points can be obtained through judgment of difference of neighborhood gray scale values and the gray scale values of the candidate feature points.

Description

technical field [0001] The invention belongs to the field of image measurement, and mainly relates to the measurement of the outer dimension of a dynamically running vehicle. Background technique [0002] In recent years, with the emphasis on the safe operation of vehicles, related products and devices for measuring vehicle outline dimensions have been continuously optimized, and detection devices such as images and lasers have appeared. At present, the traditional image detection methods usually use image stitching or overall image analysis to calculate the vehicle outline size, but the detection quality of many devices is affected by factors such as light and vehicle speed, and there are certain errors. In order to realize the accurate detection of vehicle outline information, it is necessary to optimize the image processing algorithm so that the detection device can meet the measurement of different environmental conditions and different vehicle models. Contents of the ...

AI Technical Summary

Problems solved by technology

At present, the traditional image detection methods usually use image stitching or overall image analysis to calculate the vehicle outline size, but the detection quality of many devices is affected by factors such as light and vehicle speed, and there are certain errors.

Images