CN-121994720-A - Color difference measurement method, system and storage medium based on hyperspectral global scanning imaging

CN121994720ACN 121994720 ACN121994720 ACN 121994720ACN-121994720-A

Abstract

The invention belongs to the technical field of optical detection and color measurement, and particularly relates to a color difference measurement method, a color difference measurement system and a storage medium based on hyperspectral global scanning imaging. The method comprises the steps of obtaining a hyperspectral scanning result of a to-be-detected color difference sample comprising hyperspectral scanning sub-images, splicing hyperspectral scanning images to be spliced to obtain a global hyperspectral scanning result of the to-be-detected color difference sample, enabling the hyperspectral scanning images to be spliced to be hyperspectral scanning sub-images or intermediate images formed by splicing, carrying out feature point detection on the hyperspectral scanning images to be spliced to obtain a feature point descriptor set of the hyperspectral scanning images to be spliced, carrying out matching on the feature point descriptor set of the hyperspectral scanning images to be spliced, utilizing two-dimensional rigid body coordinate transformation based on the matching result, splicing the global hyperspectral scanning images of the to-be-detected color difference sample and the standard sample, and carrying out color difference calculation on the global hyperspectral scanning images of the to be-detected color difference sample to obtain a color difference calculation result.

Inventors

DONG HAO
WANG SHU
WEI LEI
XU YUPENG
ZHANG LONG
ZHOU MINGZHU

Assignees

国家烟草质量监督检验中心
中国科学院合肥物质科学研究院

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (16)

1. The color difference measurement method based on hyperspectral global scanning imaging is characterized by comprising the following steps of: 1) Acquiring hyperspectral scanning results of a to-be-detected chromatic aberration sample comprising at least two hyperspectral scanning sub-images; 2) The hyperspectral scanning images to be spliced are spliced to obtain a global hyperspectral scanning result of the color difference sample to be detected, wherein the hyperspectral scanning images to be spliced are hyperspectral scanning sub-images or intermediate images formed by splicing; The method for stitching comprises the steps of performing feature point detection on hyperspectral scanning images to be stitched to obtain a feature point descriptor set of the hyperspectral scanning images to be stitched, matching the feature point descriptor set of the hyperspectral scanning images to be stitched, and stitching the hyperspectral scanning images to be stitched by utilizing two-dimensional rigid coordinate transformation based on a matching result; 3) And carrying out color difference calculation on the global spectrum scanning image of the color difference sample to be detected and the standard sample to obtain a color difference calculation result.
2. The method for measuring chromatic aberration based on hyperspectral global scanning imaging according to claim 1, wherein the chromatic aberration calculation result obtained in the step 3) is a chromatic aberration calculation result after filtering out the edge abnormal chromatic aberration, and the method for filtering out the edge abnormal chromatic aberration correspondingly comprises the following steps: And performing edge detection on the global hyperspectral scanning result to obtain an edge region, performing morphological expansion on the edge region to generate an edge mask so as to cover a transition region of the edge region, and filtering color difference data corresponding to the transition region from the original color difference calculation result to obtain a color difference calculation result after filtering out the edge abnormal color difference.
3. The method for measuring chromatic aberration based on hyperspectral global scanning imaging according to claim 1, wherein the specific method for performing chromatic aberration calculation on the global hyperspectral scanning result and the standard sample comprises: Converting the spliced global hyperspectral scanning result into CIEXYZ tristimulus values, and further converting the CIEXYZ tristimulus values into a color space to obtain brightness L and chromaticity a and b of each pixel point in the global hyperspectral scanning result; Matching the spliced global hyperspectral scanning result with the standard sample, determining the spatial mapping relation of the color information of each pixel point between the global hyperspectral scanning result and the standard sample based on the matching result, and calculating the color difference value between the corresponding pixel points according to the following formula: in the formula, Is the color difference between the pixel point (i, j) in the standard sample and the corresponding pixel point in the color difference sample to be detected, 、、 For standard sample pixel coordinates The luminance and chrominance parameters at which, 、、 And the brightness and chromaticity parameters of the corresponding pixel points in the color difference sample to be detected.
4. A method of color difference measurement based on hyperspectral global scanning imaging as claimed in claim 3 wherein the method of matching the stitched global hyperspectral scan result with the standard sample comprises: And respectively extracting characteristic points from the global hyperspectral scanning result and the standard sample to obtain two characteristic point descriptor sets, matching the two characteristic point descriptor sets, and determining the spatial mapping relation of the color information of each pixel point between the global hyperspectral scanning result and the standard sample through two-dimensional rigid body coordinate transformation based on the matching result.
5. The method for measuring chromatic aberration based on hyperspectral global scanning imaging according to claim 1, wherein the hyperspectral scanning result of the acquired chromatic aberration sample to be detected is the hyperspectral scanning result of the corrected chromatic aberration sample to be detected, and the correction includes black-and-white correction and reflectance calibration correction.
6. The method for measuring chromatic aberration based on hyperspectral global scanning imaging as claimed in claim 5, wherein the method for black-and-white correction comprises: obtaining a hyperspectral scanning response value of a reference white board under the condition that a light source is turned on and a hyperspectral scanning response value of the reference white board under the condition that the light source is turned off, and calculating a hyperspectral scanning result of a to-be-detected color difference sample after black and white correction by using the following formula: in the formula, And I 0 is the hyperspectral scanning response value of the to-be-detected color difference sample before black and white correction, B is the hyperspectral scanning response value of the reference white board under the condition that the light source is turned off, and W is the hyperspectral scanning response value of the reference white board under the condition that the light source is turned on.
7. The method for measuring chromatic aberration based on hyperspectral global scanning imaging as claimed in claim 5, wherein the method for reflectivity calibration correction comprises: And carrying out X-direction scanning on the standard gray plate to obtain a hyperspectral scanning response value of the standard gray plate, establishing a fitting relation between the response value and the real reflectivity of the standard gray plate, substituting the hyperspectral scanning image of the to-be-detected color difference sample into the fitting relation, and obtaining the hyperspectral scanning image of the to-be-detected color difference sample after the reflectivity calibration correction.
8. The method of hyperspectral global scanning imaging based color difference measurement as claimed in claim 5, wherein the correction further comprises a multi-region simultaneous acquisition correction for eliminating spectral noise, the method of multi-region simultaneous acquisition correction comprising: correcting a hyperspectral scanning result of a color difference sample to be detected by taking the intensity of a light source of a barium sulfate coated area on the inner wall of the integrating sphere as a reference value, wherein a correction formula is as follows: in the formula, For each pixel point (x, y) in the detection area, the wavelength of the illumination source is The corresponding hyperspectral scan response values, The barium sulfate area coated on the inner wall of the integrating sphere has the wavelength of the illumination light source And average response value data of a plurality of pixel points.
9. The method for measuring chromatic aberration based on hyperspectral global scanning imaging according to claim 2, wherein the method for edge detection is Canny edge detection.
10. The method for measuring chromatic aberration based on hyperspectral global scanning imaging as claimed in claim 1, wherein the method for matching the feature point descriptor sets of the two hyperspectral scanning sub-images is KD-tree matching.
11. The method for measuring chromatic aberration based on hyperspectral global scanning imaging as claimed in claim 10, wherein the method for screening the matching result based on the distance between the feature point and the nearest point and the next nearest point in the KD-tree matching is a we ratio screening method.
12. The method for measuring chromatic aberration based on hyperspectral global scanning imaging according to claim 1, wherein the method for feature point detection is SIFT feature point detection.
13. The method for measuring chromatic aberration based on hyperspectral global scanning imaging as claimed in claim 7, wherein the fit relationship between the response value and the true reflectivity is a quadratic function relationship with the response value as an argument.
14. The color difference measurement system based on hyperspectral global scanning imaging is characterized by comprising an optical system, a motion system and an objective table for carrying a color difference sample to be detected, wherein the motion system is used for realizing the relative movement between the optical system and the color difference sample to be detected, so that the scanning track of the optical system covers the whole color difference sample to be detected, and hyperspectral scanning results of the color difference sample to be detected are obtained; The color difference measurement system further comprises a color difference calculation module comprising a processor for implementing the color difference measurement method based on hyperspectral global scanning imaging as claimed in any one of claims 1 to 13 when executing the computer program.
15. The color difference measurement system based on hyperspectral global scanning imaging as claimed in claim 14, wherein the movement speed of the movement system is determined by the following formula: Wherein V t is the motion speed of the motion system, D is the distance between the lens of the hyperspectral camera and the sample of chromatic aberration to be detected, For the detection distance of the hyperspectral camera in the motion direction of the motion system, For the focal length of the lens of the hyperspectral camera, For the sampling frame rate of the hyperspectral camera, The number of pixels in the spatial dimension of the photosensitive element for the hyperspectral camera.
16. A computer-readable storage medium, in which a computer program is stored, characterized in that the computer program, when executed, implements the hyperspectral scanning imaging-based color difference measurement method as claimed in any one of claims 1 to 13.

Description

Color difference measurement method, system and storage medium based on hyperspectral global scanning imaging Technical Field The invention belongs to the technical field of optical detection and color measurement, and particularly relates to a color difference measurement method, a color difference measurement system and a storage medium based on hyperspectral global scanning imaging. Background Color consistency is a core indicator of package print quality assessment. The color difference tolerance requirements of high-end packaging products such as foods, cosmetics and the like are extremely high, and small color differences can influence the cognition of consumers on the quality of the products and even cause economic loss. Traditional colour difference detection mainly relies on visual colorimetry and a handheld spectrocolorimeter. In the prior art, a spectrocolorimeter, a machine vision detection method or a portable shooting and color card correction method is generally adopted to detect color difference consistency. However, the caliber of the spectrocolorimeter is usually 3-8 mm, and only the average value of the colors in a limited light spot area can be obtained, so that the color gamut distribution of the whole pattern cannot be represented. For complex patterns such as common multicolor gradual change, image-text intersection and the like in packaging printed matters, limited sampling points are difficult to cover all color areas, and the point-by-point measurement process is complicated. The method for performing millisecond-level chromaticity measurement by using a color camera has the defects of insufficient spectrum dimension and limited detection precision, and a portable shooting and color card correction scheme, such as China patent application with the application publication number of CN116559119A, provides a wood dyeing color difference detection method, a system and a medium based on deep learning. In order to solve the above problems, a detection range of a spectrocolorimeter is improved, a point-by-point scanning or image positioning-based spectrocolorimeter detection device is developed by driving the spectrocolorimeter by a motor in the prior art, but the positioning precision requirement on a motion platform is extremely high, the output is still an average color value of a selected area, and the spatial distribution characteristics of various color components in microscopic chromatic aberration and complex patterns are difficult to characterize. Furthermore, a scanning method of a line scanning hyperspectral camera is also provided in the prior art, the field of view of the method is limited by the sampling caliber of an integrating sphere, single scanning only covers a plurality of centimeters in width, the whole detection requirement of a packaged printed matter is difficult to meet, and displacement errors exist in large-size sample splicing, so that the accuracy of chromatic aberration detection is insufficient. Disclosure of Invention The invention aims to provide a color difference measurement method, a system and a storage medium based on hyperspectral global scanning imaging, which are used for solving the technical problem of low color difference detection precision in the prior art. In order to solve the technical problems, the invention provides a color difference measurement method based on hyperspectral global scanning imaging, which comprises the following steps: 1) Acquiring hyperspectral scanning results of a to-be-detected chromatic aberration sample comprising at least two hyperspectral scanning sub-images; 2) The hyperspectral scanning images to be spliced are spliced to obtain a global hyperspectral scanning result of the color difference sample to be detected, wherein the hyperspectral scanning images to be spliced are hyperspectral scanning sub-images or intermediate images formed by splicing; The method for stitching comprises the steps of performing feature point detection on hyperspectral scanning images to be stitched to obtain a feature point descriptor set of the hyperspectral scanning images to be stitched, matching the feature point descriptor set of the hyperspectral scanning images to be stitched, and stitching the hyperspectral scanning images to be stitched by utilizing two-dimensional rigid coordinate transformation based on a matching result; 3) And carrying out color difference calculation on the global spectrum scanning image of the color difference sample to be detected and the standard sample to obtain a color difference calculation result. Further, the color difference calculation result obtained in the step 3) is a color difference calculation result after filtering out the edge abnormal color difference, and the corresponding method for filtering out the edge abnormal color difference comprises the following steps: And performing edge detection on the global hyperspectral scanning result to obtain an edge region, performing morphologic