Object detection method and object detection device

Abstract

The present application discloses an object detection method and device for detecting objects in an image pair of the current frame, the image pair of the current frame includes the original image of the current frame and the corresponding disparity map of the current frame, the original image includes gray At least one of a degree map and a color map. The object detection method includes: obtaining a first disparity map detection object detected in the current frame disparity map; obtaining an original image detection object detected in the current frame original image; detecting an object according to the original image to correct the first disparity map detection object to a second disparity map detection object; and output the second disparity map detection object. Therefore, the embodiments of the present application can not only make the original image-based object detection method more robust to environmental changes, but also make the parallax image-based object detection method more accurate, even in the case of very sparse parallax.

Description

technical field [0001] The present application generally relates to the field of digital image processing, and more specifically, the present application relates to an object detection method and an object detection device. Background technique [0002] The application of object detection methods is becoming more and more popular. Specifically, in the field of computer vision, object detection and tracking has become an active research object, because it can provide key information for the computer to automatically perceive a specific object, and provide some high-level applications (such as object activity recognition and scene understanding) to provide underlying services. For example, on-road vehicle detection methods applied to driving assistance systems can effectively detect or identify the presence, distance, speed, orientation, etc. of vehicles, so as to help drivers avoid traffic accidents. [0003] Currently, there are mainly two types of object detection methods...

AI Technical Summary

Problems solved by technology

[0007] However, in this object detection method, since the grayscale features or color features are usually easily affected by environmental factors such as lighting, object detection failures may occur due to changes in the visual features of the same object in different image frames. match, and eventually cause objects to be lost in some image frames

Images