Adaptive illumination optimization method applicable to multi-reflection scene

Abstract

The invention discloses an adaptive illumination optimization method applicable to a multi-reflection scene. The adaptive illumination optimization method applicable to a multi-reflection scene comprises steps of using a projector to project four kinds grating stripes of a sine, a rectangle, a triangle and a double-frequency sine to a surface of a three-dimensional object, wherein the four kinds of grating stripes are four set of reference stripes, obtaining corresponding four sets of distorted stripes, wherein the distorted stripes are obtained by the reference stripes which are modulated by a depth of a three-dimensional object, using Fourier transform to respectively extract fundamental frequency components of four pairs of distorted stripes and reference stripes from a frequency domain, converting a folding phase which is not continuously distributed to a decompressed phase which is continuously distributed in a space through a phase decompression algorithm, establishing pixel matching between a projector and a camera, projecting a processed image to a surface of a three-dimensional target object through inverse proportion light compensation of the projector, choosing an iteration image result having a best effect for each different kind of grating projection and performing comparison of numbers of darkness / saturation pixel on four obtained adaptive illumination images to choose an optimized adaptive illumination method.

Description

technical field [0001] The invention relates to an adaptive lighting optimization method applicable to multi-reflection scenes, belonging to the field of adaptive optics; Background technique [0002] For an 8-bit digital camera, when the measured scene has different parts with high reflectivity and low reflectivity, the captured image only contains 8-bit limited pixel information. If the exposure is increased to make the part with low reflectivity clear, and The part with high reflectivity becomes saturated; conversely, if the exposure is reduced, the part with high reflectivity becomes clear, while the part with low reflectivity becomes dark, so the part with high reflectivity or low reflectivity will appear extremely bright under certain lighting conditions. Bright or extremely dark areas cause the loss of pixel information, which has an extremely adverse impact on subsequent decision-making and execution. [0003] Traditional methods mostly use manual adjustment of ligh...

AI Technical Summary

Problems solved by technology

[0002] For an 8-bit digital camera, when the measured scene has different parts with high reflectivity and low reflectivity, the captured image only contains 8-bit limited pixel information. If the exposure is increased to make the part with low reflectivity clear, and The part with high reflectivity becomes saturated; conversely, if the exposure is reduced, the part with high reflectivity becomes clear, while the part with low reflectivity becomes dark, so the part with high reflectivity or low reflectivity will appear extremely bright under certain lighting conditions. Bright or extremely dark areas cause loss of pixel information, which has an extremely adverse impact on subsequent decision-making and execution

[0003] Traditional methods mostly use manual adjustment of lighting equipment brightness to avoid overexposure. However, this manual adjustment method is not suitable for industrial automation and is not universal

Images