View-field-based monocular depth camera and object shape detection method thereof

Abstract

The invention discloses a view-field-based monocular depth camera and an object shape detection method thereof. The monocular depth camera comprises an infrared and visible compound light source, a structured light projection photoelectric engine, a 3D and 2D real-time fusing optical camera, a near infrared image sensor, a visible light image sensor and a structured light three-dimensional imagingmain control; the structured light projection photoelectric engine encodes structured light and projects the structure light to a to-be-detected object, the 3D and 2D real-time fusing optical cameracollects texture information of the surface of the to-be-detected object, the near infrared image sensor and the visible light image sensor receive 3D and 2D real-time fusing optical camera information, and the structured light three-dimensional imaging main control receives signals of the near infrared image sensor and the visible light image sensor and obtains a three-dimensional profile and texture information of the to-be-detected object. Accordingly, a parallax error of a visible light collection module and an infrared collection module in a general depth camera is eliminated, the calibration step is omitted, the measurement precision is improved, the number optical lenses needed by the general depth camera for receiving the signal is reduced to be one from two, the production cost islowered, and the system stability is improved.

Description

technical field [0001] The invention relates to the field of three-dimensional measurement and industrial detection, in particular to a field-of-view matching-based monocular depth camera and a method for detecting object topography. Background technique [0002] In the current mainstream depth camera schemes (such as structured light schemes and time-of-flight schemes), visible light images and infrared light images are generally acquired separately by two lenses, which will directly result in the visible light images and infrared light images collected by the depth camera. With parallax, which affects the accuracy of 3D reconstruction. The market needs a monocular depth camera that can omit the calibration step, improve the measurement accuracy, and reduce the optical lens group required for general depth camera signal reception from two to one. The present invention solves such problems. Contents of the invention [0003] In order to solve the deficiencies of the prior...

AI Technical Summary

Problems solved by technology

[0002] In the current mainstream depth camera schemes (such as structured light schemes and time-of-flight schemes), visible light images and infrared light images are generally acquired separately by two lenses, which will directly result in the visible light images and infrared light images collected by the depth camera. With parallax, which affects the accuracy of 3D reconstruction

Images