Robot eye adopting visual cone coordinate system for vibration zooming

Abstract

The invention relates to a robot eye. The robot eye is composed of lenses, a vibration zooming mechanism, image sensors, a vision chip and a power supply. The lenses refer to lens structures consisting of non-vibrating lens bodies. The vibration zooming mechanism is driven by an electromagnetic force to force one or more lenses to be suspended, and continuously vibrates in the direction of the optical axes of the lenses so as to realize zooming. The image sensors can be divided into two types including CCD and CMOS. The vision chip is a processor capable of converting image signals obtained bythe image sensors into machine vision, wherein the machine vision refers to stereo vision data based on a visual cone coordinate system. Moreover, the visual chip is internally provided with a lot ofvisual cone operating units for parallel computing and can discover dangerous obstacles and holes and depressions in roads, recognize transparent objects, and send a warning signal to prevent collision and fall in time.

Description

technical field [0001] The present invention relates to a robot eye; Background of the invention [0002] Currently, robot eyes use 3D depth cameras. According to different hardware implementation methods, there are currently three mainstream 3D depth cameras used in the industry: structured light, time-of-flight, and binocular stereo imaging, but the effects of these three are not comparable to human Compared with human eyes, this paper proposes a robotic eye that uses cone coordinate system vibration and zoom; Contents of the invention [0003] The robot eye using the vibratory zoom of the cone coordinate system described in this paper is composed of a lens, a vibratory zoom mechanism, an image sensor, a vision chip, and a power supply; among them, [0004] The lens refers to a lens structure, which is composed of more than one (including one) lens; the vibration zoom mechanism is driven by electromagnetic force to force more than one (including one) lens to be in a sus...

AI Technical Summary

Problems solved by technology

[0002] Currently, robot eyes use 3D depth cameras. According to different hardware implementation methods, there are currently three mainstream 3D depth cameras used in the industry: structured light, time-of-flight, and binocular stereo imaging, but the effects of these three are not comparable to human Compared with human eyes, this paper proposes a robotic eye that uses cone coordinate system vibration and zoom;

Images