RGB camera spectrum sensitivity calibration method based on spectrum-adjustable LED light source

Abstract

The invention relates to an RGB camera spectrum sensitivity calibration method based on a spectrum-adjustable LED light source, and the method comprises the following steps: S1, selecting an LED light source, and calibrating the spectrum radiation power distribution of the LED light source; S2, adjusting the distance between the emergent surface of the LED light source and the camera; S3, setting relative intensity of each spectrum channel of the LED light source and exposure parameters of the camera; S4, shooting and imaging each light source of the spectrum-adjustable LED light source to obtain RAW images; S5, extracting all pixel response values of each RGB channel according to the Bayer array arrangement of the camera; S6, averaging all pixel response values of each RGB channel to obtain a response mean value of each RGB channel; and S7, in combination with the spectral radiation power of the LED light source and the response values of the RGB three channels, calculating the spectral sensitivity of the RGB three channels of the camera through a loop optimization algorithm. The RGB camera spectrum sensitivity calibration method based on the spectrum-adjustable LED light source is simple and ordered in calibration process, high in calibration efficiency and accurate in calibration result.

Description

technical field [0001] The invention relates to a method for calibrating the spectral sensitivity of a camera, in particular to a method for calibrating the spectral sensitivity of an RGB camera based on a spectrally adjustable LED light source. Background technique [0002] The imaging process of a commercial RGB camera is as follows: the light enters the photosensitive element through the lens imaging, and after being filtered by the Bayer filter array on the photosensitive element, it is divided into three channels of red, green, and blue, and then the light at different positions After that, the camera reads the response data of all pixels (it is called a RAW image in the imaging field, which means the raw data obtained by directly reading the photosensitive element and has not been processed by the image), and finally the camera passes through the RAW image. A series of image processing operations to generate the picture RGB data presented to the user. In this process,...

AI Technical Summary

Problems solved by technology

Moreover, a smoothing term is added after the optimization algorithm, so that the estimated sensitivity function is more realistic; however, the method described in this application generally requires the measurement of the spectral power distribution of various light sources, and the reflectance of the color card to be photographed also needs to be measured in advance. Measurement requires photometry and colorimetry instruments at the same time, and the cost is high; at the same time, it is based on a large number of existing commercial camera spectral sensitivity curve databases, using PCA principal component analysis and reducing the dimension of spectral sensitivity. Although the requirements for equipment are low, However, it strongly depends on the selection of the camera spectral sensitivity curve database, and it cannot be applied to cameras with photosensitive elements with large differences in models, or cameras with non-traditional RGGB Bayer arrays

Images