CN114463291A - Shadow detection and correction method facing infrared imaging spectrometer - Google Patents

Shadow detection and correction method facing infrared imaging spectrometer Download PDF

Info

Publication number
CN114463291A
CN114463291A CN202210071512.1A CN202210071512A CN114463291A CN 114463291 A CN114463291 A CN 114463291A CN 202210071512 A CN202210071512 A CN 202210071512A CN 114463291 A CN114463291 A CN 114463291A
Authority
CN
China
Prior art keywords
shadow
image
vis
nir
swir
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210071512.1A
Other languages
Chinese (zh)
Inventor
王超
李志远
李佳
曾祥隧
许雄
童小华
谢欢
冯永玖
刘世杰
叶真
柳思聪
金雁敏
陈鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tongji University
Original Assignee
Tongji University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tongji University filed Critical Tongji University
Priority to CN202210071512.1A priority Critical patent/CN114463291A/en
Publication of CN114463291A publication Critical patent/CN114463291A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
    • G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
    • G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
    • G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/90—Dynamic range modification of images or parts thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10048—Infrared image

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Quality & Reliability (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Transforming Light Signals Into Electric Signals (AREA)

Abstract

The invention relates to a shadow detection and correction method facing an infrared imaging spectrometer, which comprises the following steps: s1, respectively performing principal component analysis and wave band combination on VIS/NIR image data, and respectively performing two normalization on a first principal component result PC1 and a wave band combination result BC 10; s2, constructing an SId image and an SIs image representing two shadow indexes by using the normalization result of the first principal component and the wave band combination; s3, performing threshold segmentation on the SID image and the SIs image respectively by using an Otsu algorithm to obtain two shadow rough detection result images; s4, further carrying out precision detection on shadow detection results of the SId and the SIs by using the absolute value of z-score to obtain a shadow precision detection result; and S5, correcting the shadow effect of the VIS/NIR and SWIR waveband spectrums by utilizing the shadow fine detection result and the spectrum data detected by the VNIS. Compared with the prior art, the method has higher detection precision on the shadow in the hyperspectral data of the infrared imaging spectrometer, and can correct the shadow effect of VIS/NIR and SWIR band spectrums.

Description

Shadow detection and correction method facing infrared imaging spectrometer
Technical Field
The invention relates to the field of spectral remote sensing data processing, in particular to a shadow detection and correction method for an infrared imaging spectrometer.
Background
Chang' e four is a moon detector launched in the second phase of China lunar exploration project and is also a detector on the back of the first landing moon of human beings; the method realizes the first soft landing and inspection tour survey of the back of the moon of human beings, has great significance and profound influence. In the many scientific payloads of Chang 'e's fourth detector year, infrared imaging spectrometer is the only scientific instrument that serves lunar mineral composition and surveys and study, can acquire lunar soil hyperspectral image and infrared spectroscopy data of millimeter level spatial resolution.
The Infrared Imaging Spectrometer (VNIS)) has a spectral range of 450-2400nm and a spectral resolution of 2-12nm, has on-orbit calibration and dust prevention functions, can adapt to temperature environments of working at-20 to +55 ℃ and storage at-50 to +70 ℃, has a weight of less than 6kg, and is a high-performance, light, small and high-integration instrument. The Infrared imaging spectrometer can image in Visible and Near-Infrared bands (VISIB and Near-Infrared, VIS/NIR, 450-.
Due to the topographic relief of the lunar surface, shadows generally exist in hyperspectral images detected by the VNIS, and due to the fact that radiation information is lost in shadow areas, the real reflectivity of the lunar surface cannot be reflected, and therefore the composition and the properties of lunar surface substances cannot be accurately analyzed. It is therefore necessary to detect and correct shadows in the image.
The existing shadow detection and correction of remote sensing images are mainly to detect earth remote sensing images by using algorithms such as edge detection, region growing and the like, but because the lunar surface texture is weak, the shadow on the lunar remote sensing images is mainly caused by terrain and micro-scale roughness, so that the shadow is irregular in shape and small in general area, and the traditional detection algorithm is difficult to be suitable for detecting lunar remote sensing images. Furthermore, in lunar geological studies, spectra around 1000nm and 2000nm are critical for inversion of mineral composition, while the detected VIS/NIR image data of infrared imaging spectrometers have wavelengths between 450nm and 950nm, and therefore shadow effects cannot be spatially corrected.
Disclosure of Invention
The invention aims to overcome the defect that a shadow detection and correction algorithm in the prior art is difficult to be applied to a moon surface hyperspectral image detected by an infrared imaging spectrometer, and provides a shadow detection and correction method for the infrared imaging spectrometer.
The purpose of the invention can be realized by the following technical scheme:
the invention provides a shadow detection and correction method facing an infrared imaging spectrometer, which comprises the following steps:
step S1: respectively performing principal component analysis and wave band combination on VIS/NIR image data, and respectively performing two normalization on a first principal component result PC1 and a wave band combination result BC 10;
step S2: constructing an SId image and an SIs image representing two shadow indexes by using a normalization result of the first principal component and the wave band combination;
step S3: respectively carrying out threshold segmentation on the SId image and the SIs image by utilizing an Otsu algorithm to obtain two shadow rough detection result images;
step S4: further carrying out precision detection on shadow detection results of the SId and the SIs by using the absolute value of z-score to obtain a shadow precision detection result;
step S5: and correcting the shadow effect of the VIS/NIR and SWIR waveband spectrums by utilizing the shadow fine detection result and the spectrum data detected by the VNIS of the infrared imaging spectrometer.
Preferably, the step S1 further includes: after performing principal component analysis on VIS/NIR image data, if the shadow is located at the negative end of the first principal component, the opposite number of the first principal component needs to be calculated, and if the shadow is located at the positive end, no change is made, and the result is denoted as PC 1.
Preferably, the band combination in step S1 is specifically:
normalizing VIS/NIR image data one by one, arranging pixel values of each normalized wave band from large to small, and respectively selecting the first 10% and the last 10% of pixels as 'non-shadow' and 'shadow';
and calculating the difference value of the pixel mean values of the 'non-shadow' and 'shadow' of each waveband, selecting the first 10 standardized wavebands with the largest difference values, accumulating the normalized wavebands, and calculating the opposite number of the accumulated result as a waveband combination result, wherein the result is recorded as BC 10.
Preferably, the band normalization expression is:
Figure BDA0003482376110000021
in the formula, BandiAnd
Figure BDA0003482376110000022
respectively representing the pixel values, mu, of the i-th band before and after normalizationiAnd σiThe pixel mean and standard deviation of the ith band are shown.
Preferably, the two normalization in step S1 are respectively:
the first is direct normalization of PC1 and BC10 to the [0,1] interval, respectively, and the results are denoted PC1d and BC10 d;
the second is to normalize the negative pixels of PC1 and BC10 to [0,1] intervals after assigning 0, and the results are denoted as PC1s and BC10 s.
Preferably, the step S2 further includes: the normalized results PC1d and BC10d are multiplied to obtain an SId image, and the normalized results PC1s and BC10s are multiplied to obtain an SIs image.
Preferably, in step S4, the refining detection process includes:
step S41: taking the non-shadow area detected by the SId image as an initial non-shadow area;
step S42: taking the shadow area detected by the SIs image as an initial shadow area;
step S43: calculating the absolute value of z-score of the shadow pixels detected by the SId image but not detected by the SIs image and the average value of the z-score of the shadow pixels detected by the SId image and the average value of the shadow pixels detected by the SIs image, wherein the absolute value is recorded as z1 and the absolute value of the z-score of the shadow pixels detected by the SIs image and the average value of the shadow pixels detected by the SIs image is recorded as z 2; and if z1 is larger than z2, marking the pixel as a shadow pixel, otherwise, marking the pixel as a non-shadow pixel, and judging to end to obtain a shadow fine detection result.
Preferably, the calculation formulas of z1 and z2 in the step S43 are expressed as follows:
Figure BDA0003482376110000031
Figure BDA0003482376110000032
in the formula, PC1*(i, j) is the value of pixel (i, j) in PC1, μfAnd σfMean and standard deviation, μ, of the unshaded areas of the SId test in PC1sAnd σsIs the mean and standard deviation of the shaded area of the SIs test in PC 1.
Preferably, in step S5, the process of correcting the shadow effect includes:
step S51: calculating the mean value of the reflectivity difference by using 10 coincident wave bands of SWIR and VIS/NIR at 900-950nm, and adding the mean value to the reflectivity of each wave band of the VIS/NIR to obtain an adjusted VIS/NIR spectrum;
step S52: calculating the proportion of shadow areas and non-shadow areas in the SWIR field of view, and calculating the average brightness values of the shadow areas and the non-shadow areas in the field of view for each wave band of the VIS/NIR image;
step S53: calculating coefficients of the average brightness reduction in the visual field caused by the shadow of each wave band of VIS/NIR according to the occupation ratio and the average brightness value of shadow areas and non-shadow areas in the SWIR visual field;
step S54: dividing the reflectivity of each wave band of VIS/NIR by the coefficient to obtain a VIS/NIR spectrum with a shadow effect corrected;
step S55: and calculating the mean value of the difference value of the reflectivity values of the VIS/NIR and the SWIR for correcting the shadow effect by using the 10 coincident wave bands of the SWIR and the VIS/NIR, and adding the mean value to the reflectivity value of each wave band of the SWIR to obtain the SWIR spectrum for correcting the shadow effect.
Preferably, the calculation formula of the coefficient in step S53 is expressed as follows:
Figure BDA0003482376110000041
in the formula, CoeffiCoefficient of the i-th band, DNsiAnd DNfiThe average brightness values of the shadow area and the non-shadow area in the SWIR field of view of the ith wave band respectively, and RATIOs and RATIOf are the proportions of the shadow and the non-shadow in the SWIR field of view respectively.
Compared with the prior art, the invention has the following advantages:
the shadow detection and correction method adopted by the invention overcomes the defect that the existing shadow detection and correction method is difficult to be applied to the moon surface hyperspectral image detected by the infrared imaging spectrometer, improves the shadow detection precision of the infrared imaging spectrometer VNIS hyperspectral data through the rough shadow detection and the fine shadow detection, improves the accuracy of the spectral data through correcting the shadow effect of VIS/NIR and SWIR wave band spectra, and is convenient for subsequent further research.
Drawings
FIG. 1 is a flow chart of a method for detecting and correcting shadows by an infrared imaging spectrometer according to the present invention;
FIG. 2 is an image corresponding to parameters generated during the algorithm implementation;
FIG. 3 is an image of the SID index and the SIs index constructed in the algorithm coarse detection stage and a shadow coarse detection result image corresponding thereto;
FIG. 4 is a comparison of the shadow result image at the algorithm fine detection stage with the original image of the embodiment;
FIG. 5 is a raw spectrum of VIS/NIR and SWIR bands and corrected for shading effects.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
Examples
As shown in fig. 1, the method for detecting and correcting a shadow of an infrared imaging spectrometer of the present invention includes the following steps:
step S1: respectively performing principal component analysis and wave band combination on VIS/NIR image data, and respectively performing two normalization on the first principal component and wave band combination results;
in step S1, a main component analysis is performed on the VIS/NIR image, and when the shadow is located at the negative end of the first main component, the inverse number of the first main component needs to be calculated, and if the shadow is located at the positive end, no transformation is performed, and the processed first main component is denoted as PC 1. Then, the VIS/NIR image data are normalized one by one, the pixel values of each normalized wave band are arranged from large to small, the first 10% and the last 10% of pixels are respectively selected as 'non-shadow' and 'shadow', the difference value of the mean values of the 'non-shadow' and 'shadow' pixels of each wave band is calculated, the first 10 normalized wave bands with the largest difference value are selected and accumulated, the opposite number of the accumulated result is calculated as a wave band combination result, and the result is recorded as BC 10. The normalized formula is as follows:
Figure BDA0003482376110000051
in the formula, BandiAnd
Figure BDA0003482376110000052
respectively representing the pixel values, mu, of the i-th band before and after normalizationiAnd σiMeans for representing the pixel mean and standard deviation of the ith band;
next, two normalization were performed for PC1 and BC10, the first was performed by directly normalizing PC1 and BC10 to the [0,1] interval, and the results were recorded as PC1d and BC10d, and the second was performed by assigning 0 to the negative value pixels of PC1 and BC10 and then normalizing to the [0,1] interval, and the results were recorded as PC1s and BC10 s.
Step S2: constructing two shadow indexes, namely SID and SIs, by utilizing the normalization result of the first main component and the wave band combination;
using PC1d, PC1S, BC10d and BC10S calculated in step S1, PC1d and BC10d are multiplied to obtain a shading index SId, and PC1S and BC10S are multiplied to obtain a shading index SIs.
Step S3: respectively carrying out threshold segmentation on the SID image and the SIs image by using an Otsu algorithm to obtain two shadow rough detection result images;
and respectively calculating a threshold value aiming at the SID index image and the SIs index image by using an Otsu algorithm, and segmenting the two index images according to respective threshold values to obtain two shadow rough detection result images.
Step S4: further carrying out precision detection on shadow detection results of the SId and the SIs by using the absolute value of z-score to obtain a shadow precision detection result;
and carrying out fine detection on the shadow by using a shadow coarse detection result obtained by detecting the SId and the SIs and an absolute value of z-score.
Step S41: taking the non-shadow area detected by the SId as an initial non-shadow area;
step S42: taking the shadow area detected by the SIs as an initial shadow area;
step S43: calculating the absolute value of z-score of the shadow pixel detected by the SID but not detected by the SIs one by one, wherein the absolute value is recorded as z1, and the absolute value of z-score of the shadow pixel detected by the SId and the mean value of the shadow pixel detected by the SIs is recorded as z2, if z1 is larger than z2, the pixel is recorded as a shadow pixel, otherwise, the pixel is recorded as a non-shadow pixel, and judging to end to obtain a shadow fine detection result;
wherein the calculation formulas of z1 and z2 are expressed as follows:
Figure BDA0003482376110000061
Figure BDA0003482376110000062
in the formula, PC1*(i, j) is the value of pixel (i, j) in PC1, μfAnd σfMean and standard deviation, μ, of the unshaded areas of the SId test in PC1sAnd σsIs the mean and standard deviation of the shaded area of the SIs test in PC 1;
step S5: correcting the shadow effect of VIS/NIR and SWIR waveband spectrums by utilizing a shadow fine detection result and spectral data detected by a VNIS;
the shadow effect of the VIS/NIR and SWIR bands is corrected using the shadow fine detection result obtained in step S4 and the VIS/NIR and SWIR band spectrum data detected by the VNIS.
Step S51: calculating the mean value of the reflectivity difference by using 10 coincident wave bands of SWIR and VIS/NIR at 900-950nm, and adding the mean value to the reflectivity of each wave band of the VIS/NIR to obtain an adjusted VIS/NIR spectrum;
step S52: calculating the proportion of shadow areas and non-shadow areas in the SWIR field of view, and calculating the average brightness values of the shadow areas and the non-shadow areas in the field of view for each wave band of the VIS/NIR image;
step S53: calculating coefficients of the average brightness reduction in the visual field caused by the shadow of each wave band of VIS/NIR according to the occupation ratio and the average brightness value of shadow areas and non-shadow areas in the SWIR visual field;
step S54: dividing the reflectivity of each wave band of VIS/NIR by the coefficient to obtain a VIS/NIR spectrum with a shadow effect corrected;
step S55: calculating the mean value of the difference values of the reflectivity of VIS/NIR and SWIR for correcting the shadow effect by utilizing 10 coincident wave bands of the SWIR and the VIS/NIR, and adding the mean value to the reflectivity of each wave band of the SWIR to obtain a SWIR spectrum for correcting the shadow effect;
wherein the calculation formula of the coefficient is expressed as follows:
Figure BDA0003482376110000071
in the formula, CoeffiCoefficient of the i-th band, DNsiAnd DNfiThe average brightness values of shadow areas and non-shadow areas in the SWIR field of view of the ith wave band, and RATIOs and RATIOf refer to the proportion of shadow and non-shadow areas in the SWIR field of view.
In this embodiment, the experiment is carried out by using 0003 image of Chang' e No. four rabbit No. two detected in the first month. Referring to fig. 2(a), the image of 760nm band is shown, and the main technical parameters of the VNIS are detailed in table 1 below.
TABLE 1
Figure BDA0003482376110000072
Referring to fig. 1, a flow chart of a shadow detection and correction algorithm for hyperspectral data of an infrared imaging spectrometer is specifically implemented as follows:
first, principal component analysis is performed on the VIS/NIR image, the first principal component image is shown in fig. 2(b), in this embodiment, the shadow is located at the negative end of the first principal component, so that the opposite number is calculated to obtain PC1, and the PC1 image is shown in fig. 2 (c). Then normalizing the VIS/NIR image data by wavebands, arranging the normalized pixel values of each waveband from large to small, selecting the first 10% and the last 10% of the pixels as "non-shadow" and "shadow", calculating the difference between the average values of the pixels in each waveband "non-shadow" and "shadow", and calculating the difference between the average values of the pixels in each waveband "non-shadow" and "shadow", wherein the difference is (3.3148,3.3047,3.2656,3.2924,3.2946,3.2464,3.271,3.2704,3.2735,3.2997,3.2872,3.2441,3.2593,3.2835,3.251,3.2356,3.26,3.2606,3.2533,3.2732,3.3001,3.2799,3.265,3.284,3.321,3.3045,3.2733,3.2698,3.2984,3.3044,3.2824,3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 68538, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 6854,6854, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 68538, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 3.2687, 68538, 3.2687, 3.2687, 3.2687, 3.2687, 3.3286,3.3367,3.3447,3.3363,3.3414,3.3332,3.3254,3.3132,3.306,3.3018,3.3095,3.318,3.3274,3.3355,3.3432,3.3503,3.3532,3.3509,3.3397,3.3376,3.3322). The first 10 normalized bands with the largest difference are selected, the band numbers of the ten bands are (40,75,74,39,76,41,73,96,97,95), the normalized pixel values of the ten bands are accumulated, and the inverses of the normalized pixel values are calculated to obtain band combination results BC10, BC10, see fig. 2 (d).
Two normalization were performed on PC1 and BC10, respectively, to yield PC1d, BC10d, PC1s, and BC10 s. Images of PC1d, BC10d, PC1s, and BC10s are shown in fig. 2(e), (f), (g), and (h).
PC1d and BC10d are multiplied to obtain SID, and the SId image is shown in FIG. 3 (a). PC1s and BC10s are multiplied to obtain the SIs, and the SIs image is shown in FIG. 3 (b).
And calculating threshold segmentation SId and SIs by using an Otsu algorithm to obtain an SId shadow coarse detection result and an SIs shadow coarse detection result. The threshold for the segmentation SId was calculated to be 0.4782, and the result of coarse detection of SId shadow is shown in fig. 3 (c). The threshold for the segmentation of the SIs was calculated to be 0.2471, and the coarse detection result of the SIs shadow is shown in fig. 3 (d).
And carrying out fine detection on the shadow by using a shadow coarse detection result obtained by detecting the SId and the SIs and an absolute value of z-score. Firstly, the non-shadow area detected by the SId is used as an initial non-shadow area, and the shadow area detected by the SIs is used as an initial shadow area. Then, for the shaded pixels detected by SId but not detected by SIs, the absolute value of z-score with the mean of the non-shaded regions detected by SId is calculated one by one to obtain z1, and the absolute value of z-score with the mean of the shaded regions detected by SIs to obtain z 2. Here, taking the pixels in the 97 th row and the 122 th column as an example, the pixels are taken as shadow pixels because z1 is 4.2362, z2 is 1.8817, and z1 is larger than z 2. And calculating the shadow pixels detected by the SID but not detected by the SIs one by one until all shadow pixels are calculated, and obtaining a shadow fine detection result, which is shown in FIG. 4 (b).
And correcting the shadow effect of the VIS/NIR and SWIR bands by using the shadow fine detection result and spectral data of the VIS/NIR and SWIR bands detected by the VNIS.
1) The average value of the reflectivity difference is calculated by using 10 coincident wave bands of SWIR and VIS/NIR at 900-950nm, the calculation result in the embodiment is 0.0089, and the reflectivity of each wave band of the VIS/NIR is added to 0.0089 to obtain the adjusted VIS/NIR spectrum, which is shown in FIG. 5.
2) (iv) calculating the ratio of shadow areas in the SWIR field of view, ratioos (r) 0.3647) and non-shadow areas, ratioos (r) 0.6353, calculating the average luminance value of shadow areas and non-shadow areas in the field of view for each wavelength band of VIS/NIR images, (0.0047092,0.0035426,0.0030663,0.0032207,0.0029567,0.0025,0.0028319,0.0026762,0.0031372,0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, 0.00325, the average luminance value of the unshaded region is (,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, etc.).
3) Calculating the coefficient Coeff of the reduction of the average brightness in the visual field caused by the shadow in each VIS/NIR wave bandiIs (0.77856,0.75908,0.74195,0.74916,0.74915,0.7302,0.73829,0.73969,0.74807,0.75155,0.75449,0.73041,0.73397,0.74788,0.73292,0.73199,0.73172,0.7302,0.73495,0.73741,0.7486,0.7421,0.73877,0.74706,0.7602,0.75483,0.74648,0.73901,0.75045,0.74891,0.74141,0.74094,0.74711,0.74247,0.74116,0.73455,0.74012,0.77007,0.785,0.79757,0.79074,0.7738,0.76086,0.76349,0.77206,0.7677,0.75505,0.74954,0.7486,0.75313,0.7536,0.7565,0.758,0.74804,0.74581,0.75084,0.75667,0.75718,0.76409,0.76217,0.75625,0.75565,0.76027,0.76799,0.7751,0.78091,0.77874,0.77955,0.77647,0.77431,0.77452,0.77634,0.78207,0.79313,0.79481,0.79394,0.79202,0.78807,0.7829,0.77839,0.78118,0.79079,0.78725,0.78915,0.78431,0.78102,0.77819,0.77445,0.77155,0.77653,0.78364,0.79095,0.80017,0.80377,0.80353,0.8047,0.80026,0.79934,0.79431,0.79365)
4) Reflectivity for each of the VIS/NIR bands divided by the corresponding coefficient CoeffiVIS/NIR spectra corrected for shadow effects were obtained, see FIG. 5.
5) The average value of the difference between the reflectances of the VIS/NIR and the SWIR for correcting the shadowing effect is calculated by using 10 coincident bands of the SWIR and the VIS/NIR, the calculation result in this embodiment is 0.0143, and the reflectivity of each band of the SWIR is added to 0.0143 to obtain the SWIR spectrum for correcting the shadowing effect, which is shown in fig. 5.
Fig. 4(a) shows an original image of the image in the 760nm band of the present embodiment, fig. 4(b) shows a shadow detection result, and it can be seen by comparing the two images that the shadow detection precision of the present algorithm is high, and it can be seen by looking at fig. 5 that the present algorithm can correct the shadow effect of the VIS/NIR and SWIR band spectra.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (10)

1. A shadow detection and correction method facing an infrared imaging spectrometer is characterized by comprising the following steps:
step S1: respectively performing principal component analysis and wave band combination on VIS/NIR image data, and respectively performing two normalization on a first principal component result PC1 and a wave band combination result BC 10;
step S2: constructing an SId image and an SIs image representing two shadow indexes by using a normalization result of the first principal component and the wave band combination;
step S3: respectively carrying out threshold segmentation on the SId image and the SIs image by utilizing an Otsu algorithm to obtain two shadow rough detection result images;
step S4: further carrying out precision detection on shadow detection results of the SId and the SIs by using the absolute value of z-score to obtain a shadow precision detection result;
step S5: and correcting the shadow effect of the VIS/NIR and SWIR waveband spectrums by utilizing the shadow fine detection result and the spectrum data detected by the VNIS of the infrared imaging spectrometer.
2. The method for detecting and correcting shadow of infrared imaging spectrometer as claimed in claim 1, wherein said step S1 further comprises: after the VIS/NIR image data is subjected to principal component analysis, if the shadow is located at the negative end of the first principal component, the inverse of the first principal component needs to be calculated, and if the shadow is located at the positive end, no change is made, and the result is denoted as PC 1.
3. The method for detecting and correcting the shadow facing the infrared imaging spectrometer as claimed in claim 1, wherein the combination of the wave bands in the step S1 is specifically:
normalizing VIS/NIR image data one by one, arranging pixel values of each normalized wave band from large to small, and respectively selecting pixels of the first 10% and the last 10% as 'non-shadow' and 'shadow';
and calculating the difference value of the pixel mean values of the 'non-shadow' and 'shadow' of each waveband, selecting the first 10 standardized wavebands with the largest difference values, accumulating the normalized wavebands, and calculating the opposite number of the accumulated result as a waveband combination result, wherein the result is recorded as BC 10.
4. The method of claim 3, wherein the band normalization expression is:
Figure FDA0003482376100000011
in the formula, BandiAnd
Figure FDA0003482376100000012
respectively representing the pixel values, mu, of the i-th band before and after normalizationiAnd σiThe pixel mean and standard deviation of the ith band are shown.
5. The method for detecting and correcting shadow of infrared imaging spectrometer as claimed in claim 1, wherein two normalization in step S1 are respectively:
the first is direct normalization of PC1 and BC10 to the [0,1] interval, respectively, and the results are denoted PC1d and BC10 d;
the second is to normalize the negative pixels of PC1 and BC10 to [0,1] intervals after assigning 0, and the results are denoted as PC1s and BC10 s.
6. The method for detecting and correcting shadow of infrared imaging spectrometer as claimed in claim 5, wherein said step S2 further comprises: the normalized results PC1d and BC10d are multiplied to obtain an SId image, and the normalized results PC1s and BC10s are multiplied to obtain an SIs image.
7. The method for detecting and correcting shadow of infrared imaging spectrometer as claimed in claim 1, wherein in step S4, the step of refining the detection comprises:
step S41: taking the non-shadow area detected by the SId image as an initial non-shadow area;
step S42: taking the shadow area detected by the SIs image as an initial shadow area;
step S43: calculating the absolute value of z-score of the shadow pixels detected by the SId image but not detected by the SIs image and the average value of the z-score of the shadow pixels detected by the SId image and the average value of the shadow pixels detected by the SIs image, wherein the absolute value is recorded as z1 and the absolute value of the z-score of the shadow pixels detected by the SIs image and the average value of the shadow pixels detected by the SIs image is recorded as z 2; and if z1 is larger than z2, marking the pixel as a shadow pixel, otherwise, marking the pixel as a non-shadow pixel, and judging to end to obtain a shadow fine detection result.
8. The method of claim 7, wherein the equations for z1 and z2 in step S43 are as follows:
Figure FDA0003482376100000021
Figure FDA0003482376100000022
in the formula, PC1*(i, j) is the value of pixel (i, j) in PC1, μfAnd σfMean and standard deviation, μ, of the unshaded areas of the SId test in PC1sAnd σsIs the mean and standard deviation of the shaded area of the SIs test in PC 1.
9. The method for detecting and correcting shadow of an infrared imaging spectrometer as claimed in claim 1, wherein the step S5, correcting the shadow effect comprises:
step S51: calculating the mean value of the reflectivity difference by using 10 coincident wave bands of SWIR and VIS/NIR at 900-950nm, and adding the mean value to the reflectivity of each wave band of the VIS/NIR to obtain an adjusted VIS/NIR spectrum;
step S52: calculating the proportion of shadow areas and non-shadow areas in the SWIR field of view, and calculating the average brightness values of the shadow areas and the non-shadow areas in the field of view for each wave band of the VIS/NIR image;
step S53: calculating coefficients of the average brightness reduction in the visual field caused by the shadow of each wave band of VIS/NIR according to the occupation ratio and the average brightness value of shadow areas and non-shadow areas in the SWIR visual field;
step S54: dividing the reflectivity of each wave band of VIS/NIR by the coefficient to obtain a VIS/NIR spectrum with a shadow effect corrected;
step S55: and calculating the mean value of the difference value of the reflectivity values of the VIS/NIR and the SWIR for correcting the shadow effect by using the 10 coincident wave bands of the SWIR and the VIS/NIR, and adding the mean value to the reflectivity value of each wave band of the SWIR to obtain the SWIR spectrum for correcting the shadow effect.
10. The method for detecting and correcting shadow of infrared imaging spectrometer as claimed in claim 9, wherein the calculation formula of the coefficients in step S53 is represented as follows:
Figure FDA0003482376100000031
in the formula, CoeffiCoefficient of the i-th band, DNsiAnd DNfiThe average brightness values of the shadow area and the non-shadow area in the SWIR field of view of the ith wave band respectively, and RATIOs and RATIOf are the proportions of the shadow and the non-shadow in the SWIR field of view respectively.
CN202210071512.1A 2022-01-21 2022-01-21 Shadow detection and correction method facing infrared imaging spectrometer Pending CN114463291A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210071512.1A CN114463291A (en) 2022-01-21 2022-01-21 Shadow detection and correction method facing infrared imaging spectrometer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210071512.1A CN114463291A (en) 2022-01-21 2022-01-21 Shadow detection and correction method facing infrared imaging spectrometer

Publications (1)

Publication Number Publication Date
CN114463291A true CN114463291A (en) 2022-05-10

Family

ID=81409525

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210071512.1A Pending CN114463291A (en) 2022-01-21 2022-01-21 Shadow detection and correction method facing infrared imaging spectrometer

Country Status (1)

Country Link
CN (1) CN114463291A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115410096A (en) * 2022-11-03 2022-11-29 成都国星宇航科技股份有限公司 Satellite remote sensing image multi-scale fusion change detection method, medium and electronic device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115410096A (en) * 2022-11-03 2022-11-29 成都国星宇航科技股份有限公司 Satellite remote sensing image multi-scale fusion change detection method, medium and electronic device
CN115410096B (en) * 2022-11-03 2023-01-24 成都国星宇航科技股份有限公司 Satellite remote sensing image multi-scale fusion change detection method, medium and electronic device

Similar Documents

Publication Publication Date Title
Giannini et al. Land Surface Temperature from Landsat 5 TM images: comparison of different methods using airborne thermal data.
US6909815B2 (en) Method for performing automated in-scene based atmospheric compensation for multi-and hyperspectral imaging sensors in the solar reflective spectral region
Gong et al. Forest canopy closure from classification and spectral unmixing of scene components-multisensor evaluation of an open canopy
Kruse Mineral mapping with AVIRIS and EO-1 Hyperion
US7184890B2 (en) Cloud shadow detection: VNIR-SWIR
Collings et al. Techniques for BRDF correction of hyperspectral mosaics
Pandya et al. Bandpass solar exoatmospheric irradiance and Rayleigh optical thickness of sensors on board Indian remote sensing satellites-1B,-1C,-1D, and P4
CN111553922B (en) Automatic cloud detection method for satellite remote sensing image
US6778702B1 (en) Method and apparatus for assessing the quality of spectral images
Collings et al. Empirical models for radiometric calibration of digital aerial frame mosaics
CN114463291A (en) Shadow detection and correction method facing infrared imaging spectrometer
Settle On constrained energy minimization and the partial unmixing of multispectral images
Deng et al. Estimating chlorophyll content of apple leaves based on preprocessing of reflectance spectra
Miglani et al. Evaluation of EO-1 Hyperion data for agricultural applications
Ashton et al. A novel method for illumination suppression in hyperspectral images
CN109696406B (en) Moon table hyperspectral image shadow region unmixing method based on composite end member
MOHAMMADY et al. Role of smile correction in mineral detection on hyperion data
Katkovsky Base of Surface Spectral Characteristics from Photospectral Measurements on the International Space Station.
Isnaen et al. Comparison of object spectral reflectance from WorldView-2 image and field measurement
Nauss et al. Package ‘satellite’
Vincent et al. Subpixel target detection in hyperspectral imaging
US9159138B2 (en) Device and method for dynamic adaptation of spatial resolution for imager fourier transform spectrometers
Douglas A novel endmember, fractional abundance, and contrast model for hyperspectral imagery
Chandra et al. Using coupled subspace models for reflectance/illumination separation
Ientilucci et al. Tracking of vehicles across multiple radiance and reflectance hyperspectral datasets

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination