A Method for Accurate Registration of RGB and Depth Information

Abstract

The embodiment of the present invention discloses a method for precise registration of RGB and depth information, including preparing a depth sensor and an RGB sensor device, preprocessing the depth sensor, setting a synchronous shooting signal, building a semi-automatic checkerboard data acquisition platform, synchronizing Take the depth image and RGB image of the checkerboard for calibration, and calculate the corresponding pixel coordinates of the RGB image (i y , j y ), through the calculated IR space coordinates, RGB space coordinates, and RGB pixel values, select and output the desired coordinate data through the selector. Through the solution of the present application, accurate registration of RGB and depth information can be realized.

Description

technical field [0001] The invention belongs to the technical fields of image processing, natural interaction and integrated circuits, and in particular relates to a method and device for precise registration of RGB and depth information (Depth). Background technique [0002] Depth perception is one of the important research directions in the field of computer vision, which enables smart devices to obtain capabilities similar to human visual perception. Existing depth perception techniques can be divided into active methods and passive methods. The depth perception technology based on binocular stereo vision is a passive method, which is greatly affected by ambient light and the algorithm and hardware implementation are complex. Active methods mainly include structured light method (coding light) and time-of-flight method (time offilght). Among them, the depth perception technology based on the structured light method can obtain more accurate depth information, and is not ...

AI Technical Summary

Problems solved by technology

However, there are more or less problems in the existing camera calibration methods. For example, the traditional camera calibration method has low calibration accuracy; the nonlinear calibration method has complex algorithms and is difficult to achieve real-time calibration; The camera calibration method is convenient and fast, but the depth image directly captured by the depth sensor cannot generate the characteristics of the checkerboard calibration board, and cannot be directly used for calibration

Images