Image processing method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides an image processing method and device, electronic equipment, a computer readable storage medium and a computer program product, and is applied to the field of maps. The method comprises the following steps: respectively performing target identification processing on a first image and a second image to obtain a plurality of detection frames in the first image and a plurality of detection frames in the second image; for the first image and the second image, alternately performing fusion processing of different detection frames in the same image and fusion processing of different detection frames in different images to obtain features of each detection frame in the first image and features of each detection frame in the second image; determining the similarity between any one detection frame in the first image and any one detection frame in the second image based on the feature of each detection frame in the first image and the feature of each detection frame in the second image; and generating an object recognition result according to the similarity. According to the invention, the matching precision and efficiency of the to-be-recognized objects in different images can be improved.

Description

technical field [0001] The present application relates to the technical fields of the Internet and intelligent transportation, and in particular to an image processing method, device, electronic equipment, and computer-readable storage medium. Background technique [0002] With the development of location services and navigation applications, whether real-time and accurate road network data can be provided has become an important indicator to measure the level of map services. For example, in a pair of road scene images with similar shooting locations but different shooting times and shooting angles, find objects that appear at the same time (such as traffic elements, including speed limit signs, traffic limit signs, traffic lights, number of lanes, road paving status, etc.) ), is a very important task. In the fields of vision-based automatic update of road data, high-precision maps, and automatic driving, this technology can be relied on to reduce the workload of manual ve...

AI Technical Summary

Problems solved by technology

[0003] However, the solutions provided by related technologies usually use key point detection and matching methods to calculate projection matrices and other artificial strategy-based solutions to estimate whether the objects appearing in two images are the same. This solution is very dependent on human experience, and there are Problems such as low matching efficiency, inflexibility, and difficulty in adjustment

Images