Three-dimensional depth perception method and three-dimensional depth perception device based on adjustable working range

Abstract

The invention discloses a three-dimensional depth perception method and a three-dimensional depth perception device based on an adjustable working range. The three-dimensional depth perception method comprises the following steps: setting a working range mode in an external setting or self-adaptive adjusting manner, projecting a coded pattern in a corresponding working range by adjusting the driving current of a driving circuit of a laser pattern projector, adjusting the focal length and the parallax range of a receiving camera, collecting a projected coded image sequence and feeding into a depth perception module, adjusting pre-processing image control parameters by the depth perception module according to the working range mode, performing block matching parallax calculation and depth calculation of the input coded image sequence by selecting a set of reference coded images conforming to the working range mode, and outputting a depth image sequence. On the basis of the method, the three-dimensional depth perception device based on the adjustable working range is realized; the device is beneficial to optimizing and realizing the depth perception calculation method in real time; therefore, the depth perception calculation method is applicable to different working ranges; and the application limitation of the existing three-dimensional depth perception device is broken through.

Description

technical field [0001] The disclosure relates to the technical fields of image processing, natural interaction and integrated circuits, and in particular to a three-dimensional depth perception method and device with an adjustable working range. Background technique [0002] Natural and harmonious human-computer interaction is the ideal goal of human beings to control machines, so that machines can understand the commands transmitted by people in a natural state. Among them, depth perception technology is the core technology of natural human-computer interaction. In machine vision, intelligent monitoring , 3D reconstruction, somatosensory interaction, 3D printing, drones and other fields have broad application prospects. The active vision mode based on structured light can obtain the depth information of the image more accurately, such as projecting a fixed-pattern image onto the surface of an object for encoding through an infrared laser, collecting an infrared encoded imag...

AI Technical Summary

Problems solved by technology

These existing three-dimensional depth sensing devices cannot realize adjustable working ranges. For example, the same set of devices can be used for close-range gesture or face recognition, and can also be used for somatosensory action recognition several meters away, or for longer distances (such as 10 Pedestrian tracking and recognition

Images