Automatic exposure method and system of differential absorption spectrometer

Abstract

The invention discloses an automatic exposure method and system of a differential absorption spectrometer, and relates to the technical field of spectrometers, and parameters are set based on a calibrated rapid exposure differential imaging spectrum system; under the current light intensity, hyperspectral data output by the differential absorption spectrometer and three-channel numerical values output by the visual camera are obtained at the same time, and an exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity is obtained based on the three-channel numerical values and set parameters; substituting the target pixel value of the hyperspectral data into an exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity to obtain target exposure time; correcting the target exposure time, setting the corrected target exposure time as the exposure time of the differential absorption spectrometer, and feeding back the corrected target exposure time to the differential absorption spectrometer to obtain spectral data; through the automatic exposure method and system, hyperspectral data with a high signal-to-noise ratio can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of spectrometers, and in particular to an automatic exposure method and system for a differential absorption spectrometer. Background Art

[0002] In the field of spectral imaging technology, in order to ensure data acquisition efficiency and spectral data quality, it is necessary to set different exposure times according to the changes in light intensity in actual application scenarios.

[0003] Traditional automatic exposure methods are not suitable for spectrometer-type imaging systems, and usually require at least 3-4 images for exposure measurement, wasting the data acquisition time window; multiple imaging using the spectral data obtained by the imaging spectrometer can obtain automatic exposure values, but this is no longer applicable to imaging spectrometers that use mechanical shutters or situations that require higher acquisition efficiency.

[0004] The imaging brightness adjustment parameters of the imaging system usually include three types: adjusting the aperture size, adjusting the exposure time, and adjusting the imaging gain (i.e., the magnification). Therefore, the method of adjusting the brightness of the shooting target is to adjust these three parameters.

[0005] Automatic exposure means that the imaging system automatically adjusts the system parameters so that the target brightness reaches the specified standard. This standard is the evaluation algorithm for the target brightness, which usually has three types: average brightness method, weighted average method and brightness histogram. The average brightness method is to average the brightness of all pixels in the image and finally reach the target brightness by continuously adjusting the exposure parameters. The weighted average method is to set different weights for different areas of the image to calculate the image brightness. The brightness histogram method calculates the image brightness by assigning different weights to the peaks in the histogram.

[0006] The spectrometer imaging system obtains the spectral information of the target area, and the average brightness method can be used to evaluate the target brightness (i.e., photometry). This method obtains the average brightness information of the entire band of the target area. When it is necessary to obtain high brightness information of a certain characteristic band, the data of the corresponding inversion band can be obtained, so that the specified band has a higher brightness, that is, the band-related automatic exposure imaging method. This method obtains the target light intensity by using multiple exposures and calculates the target exposure time, and shortens the photometry time by reading only a few lines of data from the detector.

[0007] Differential imaging spectrometers usually invert the spectrum in the 300-400nm band to obtain the trace gas concentration in the target area. In order to fully improve the acquisition efficiency and spectral data quality, a fast and automatic exposure imaging method is required.

[0008] Existing photometry methods are not suitable for imaging spectrometers: (1) The photometry time is long; usually, 3-4 full-image images need to be taken for photometry before data acquisition, which wastes the data acquisition time window. (2) Band-dependent automatic exposure imaging methods still need to occupy the acquisition window time for multiple exposures and imaging, and are not applicable to occasions where mechanical shutters are used or where more efficient data acquisition is required. Summary of the invention

[0009] Based on the technical problems existing in the background technology, the present invention proposes an automatic exposure method and system for a differential absorption spectrometer, which does not require the use of a differential absorption spectrometer for light measurement, saves the light measurement cycle, and makes full use of the data acquisition time window, which can greatly improve the data acquisition efficiency of the equipment, realize fast automatic exposure imaging, and extend the service life of the mechanical shutter.

[0010] The present invention provides an automatic exposure method for a differential absorption spectrometer, comprising the following steps:

[0011] Step 1: Set the exposure time, aperture value and fitting parameters of the visible camera based on the calibrated fast exposure differential imaging spectroscopy system, and set the exposure time range [T min , T cali ], and also obtain the original exposure time-pixel value curve corresponding to the differential absorption spectrometer;

[0012] Step 2: Under the current light intensity, the hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visual camera are simultaneously obtained, and the differential absorption spectrometer is calculated based on the three-channel values ​​and the fitting parameters at the exposure time T cali The pixel value under the current light intensity is combined with the bias value corresponding to the 0-second exposure in the original exposure time-pixel value curve to obtain the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity;

[0013] Step 3: Substitute the target pixel value of the hyperspectral data into the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity to obtain the target exposure time T;

[0014] Step 4: Determine whether the target exposure time T is within the exposure time range [T min , T cali ] will be less than the minimum exposure time T min The target exposure time T is set to T = T min , will be greater than the maximum exposure time T cali The target exposure time T is set to T = T cali , to correct the target exposure time T to the exposure time range [T min , T cali ];

[0015] Step 5: Set the corrected target exposure time T as the exposure time of the differential absorption spectrometer and feed it back to the differential absorption spectrometer to obtain spectral data.

[0016] Furthermore, the calibration process of the rapid exposure differential imaging spectroscopy system is as follows:

[0017] (a1) Determine the target light intensity range of the differential absorption spectrometer based on the actual application scenario [L min , L max ] and the original exposure time-pixel value curve of the differential absorption spectrometer under different illumination, L min is the minimum light intensity value, L max is the maximum light intensity value;

[0018] (a2) According to the noise level of the differential absorption spectrometer imaging, the noise area Noise Zone1 of the maximum output value of the differential absorption spectrometer is obtained, so as to set the pixel value boundary condition [DN min , D.N. max ] and exposure time range [T min , T cali ];

[0019] (a3) According to the light intensity range of the observed target [L min , L max ] Determine the exposure time-grayscale value relationship curve of the visible camera at different apertures, and set the effective grayscale value range of the visible camera imaging in combination with the noise area Noise Zone2 of the visible camera [Gr min , Gr max ];

[0020] (a4) Set the exposure time T2 of the visible camera to be less than the maximum exposure time T cali , the exposure time-grayscale value relationship curve under different apertures under the exposure time T2 is obtained, and the effective grayscale value range closest to the effective grayscale value range [Gr min , Gr max ] is used as the effective aperture, at which point the dynamic range of the visible camera is the largest;

[0021] (a5) setting the exposure time and aperture size of the visual camera based on the exposure time T2 and the effective aperture;

[0022] (a6) based on the fact that the field of view of the visible camera covers the field of view of the differential absorption spectrometer, calibrating the field of view of the visible camera and the field of view of the differential absorption spectrometer;

[0023] (a7) Using the calibrated visible camera and differential absorption spectrometer, the target in the actual application scene is photographed to obtain the light intensity range of multiple observation targets [Lmin , L max ] The hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visual camera;

[0024] (a8) Based on the field of view in (a6), the common field of view data of the differential absorption spectrometer and the visible camera, the pixel value boundary conditions [DN min , D.N. max ] and the effective gray value range [Gr min , Gr max ], the three-channel numerical and hyperspectral data are fitted to obtain the relationship between the pixel values ​​of the differential absorption spectrometer and the pixel values ​​of the visible camera, and the fitting parameters are obtained, which are the calibration results.

[0025] Furthermore, in (a2), the pixel value boundary condition of the differential absorption spectrometer during the calibration phase [DN min , D.N. max ] and exposure time range [T min , T cali The setting process of ] is as follows:

[0026] Set the maximum effective pixel value DN max Below 50% of the noise area Noise Zone1, the maximum effective pixel value DN is obtained max Corresponding to the maximum light intensity condition L max The maximum exposure time T cali ;

[0027] Combined with the minimum light intensity value L min The original exposure time-pixel value curve under the exposure time T cali At the minimum light intensity L min The minimum valid pixel value DN corresponding to min ;

[0028] According to the maximum light intensity condition L max The original exposure time-pixel value curve under the condition of min The corresponding time is set as the minimum exposure time T min .

[0029] Furthermore, in (a3), the effective grayscale value range of the visible camera imaging [Gr min , Gr max The setting process of ] is as follows:

[0030] Set the maximum effective grayscale value Gr max The maximum effective gray value Gr under different apertures is obtained by setting it to 50% of the noise area Noise Zone2 of the visible camera. max Corresponding to the maximum light intensity condition L maxThe maximum exposure time T s ;

[0031] Combined with the minimum light intensity value L at different apertures min The corresponding exposure time-gray value relationship curve is obtained, and the exposure time T s At the minimum light intensity L min The minimum effective gray value Gr min .

[0032] Further, in (a8): the three-channel numerical value and the hyperspectral data are fitted to obtain the relationship between the pixel value of the differential absorption spectrometer and the pixel value of the visible camera, and the fitting parameters are obtained, wherein the relationship formula between the pixel value of the differential absorption spectrometer and the pixel value of the visible camera is as follows:

[0033]

[0034] Among them, N is the pixel value of the differential absorption spectrometer, α, β, γ, α′, β′, and γ′ are fitting parameters, R, G, and B are the three-channel values ​​of the visible camera, and δ and δ′ are balance parameters.

[0035] Further, in step 2, based on the three-channel values ​​and the fitting parameters and using the relationship formula between the pixel value of the differential absorption spectrometer and the pixel value of the visible camera, the differential absorption spectrometer at the exposure time T is calculated. cali The pixel value below.

[0036] An automatic exposure system for a differential absorption spectrometer realizes automatic exposure of the differential absorption spectrometer based on a calibrated rapid exposure differential imaging spectrometer system, wherein the rapid exposure differential imaging spectrometer system comprises a differential absorption spectrometer, a visual camera and a processor;

[0037] The calibrated rapid exposure differential imaging spectroscopy system specifically performs the following process:

[0038] Step 1: Set the exposure time, aperture value and fitting parameters of the visible camera based on the calibrated fast exposure differential imaging spectroscopy system, and set the exposure time range [T min , T cali ], and also obtain the original exposure time-pixel value curve corresponding to the differential absorption spectrometer;

[0039] Step 2: Under the current light intensity, the processor simultaneously obtains the hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visible camera, and calculates the differential absorption spectrometer at the exposure time T based on the three-channel values ​​and the fitting parameters. caliThe pixel value under the current light intensity is combined with the bias value corresponding to the 0-second exposure in the original exposure time-pixel value curve to obtain the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity;

[0040] Step 3: The processor brings the target pixel value of the hyperspectral data into the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity to obtain the target exposure time T;

[0041] Step 4: The processor determines whether the target exposure time T is within the exposure time range [T min , T cali ] will be less than the minimum exposure time T min The target exposure time T is set to T = T min , will be greater than the maximum exposure time T cali The target exposure time T is set to T = T cali , to correct the target exposure time T to the exposure time range [T min , T cali ];

[0042] Step 5: The processor sets the corrected target exposure time T as the exposure time of the differential absorption spectrometer and feeds it back to the differential absorption spectrometer to obtain spectral data.

[0043] The advantages of the automatic exposure method and system of a differential absorption spectrometer provided by the present invention are as follows: a fast automatic exposure imaging method and a dual imaging system suitable for a differential absorption spectrometer are proposed, and the visible camera system is cheap and easy to obtain, and the overall cost is low; by prior drawing the exposure time-pixel value curve (i.e., T-DN curve) of the imaging spectrometer, combined with the calculated spectral imaging pixel value (DN value) under the current light intensity, the exposure time corresponding to the expected DN value can be directly obtained, and the calculation is efficient and fast; in addition, a differential imaging spectrometer automatic exposure calibration method is proposed, combined with the actual application scenario, the differential absorption spectrometer and the visible camera RGB three-channel data are obtained, and the fitting parameters are obtained by the piecewise function fitting method; the fitting parameters are obtained by system calibration, and combined with the working principles of the imaging spectrometer and the visible camera imaging system, a method for drawing the exposure time-pixel value curve (T-DN curve) and the exposure time-gray value relationship curve (T-Gr curve) is proposed, which can make full use of the dynamic range of the two imaging systems and improve the system calibration accuracy. The frame rate of the visible camera set at the same time is higher than that of the imaging spectrometer, which can realize real-time adjustment of the exposure time parameters of the differential absorption spectrometer and improve the data acquisition efficiency of the observation time window. Moreover, there is no need for multiple exposure imaging, which reduces the number of imaging times of the differential absorption spectrometer and can greatly increase the shutter service life of the mechanical shutter imaging spectrometer. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1It is a structural schematic diagram of the rapid exposure differential imaging spectroscopy system of the present invention;

[0045] Figure 2 This is a flow chart of the calibrated rapid exposure differential imaging spectroscopy system in the application phase;

[0046] Figure 3 Schematic diagram of the exposure calculation method of the fast exposure differential imaging spectral system. (a) is the T-DN curve of the differential absorption spectrometer imaging system; (b) is the T-Gr curve of the visible camera imaging system; (c) is the T-Gr curve corresponding to different aperture sizes of the visible camera imaging system under a specific exposure time; (d) is a schematic diagram of the pixel value of the spectrometer imaging system obtained by mapping the pixel values ​​of the RGB channels of the visible camera. DETAILED DESCRIPTION

[0047] Below, the technical solution of the present invention is described in detail through specific embodiments. Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific implementation disclosed below.

[0048] In order to optimize the photometry time of the imaging spectrometer, this embodiment adopts a dual imaging system mode, that is, adding an ordinary RGB color visual camera to the differential absorption spectrometer. Currently, the RGB visual camera on the market is cheap and has excellent imaging quality, so the cost is low. System block diagram, such as Figure 1 As shown. In the figure, the differential absorption spectrometer collects spectrometer data, and the visual camera collects spectral data of three channels of RGB. The data receiving and processing computer (i.e., the processor) simultaneously receives the hyperspectral data of the imaging spectrometer and the color image data of the visual camera. This embodiment uses the color image data of the visual camera to calculate the imaging exposure time parameters of the differential absorption spectrometer in real time, thereby realizing the automatic exposure of the differential imaging spectrometer.

[0049] It should be noted that this embodiment is divided into a calibration phase and an application phase. The calibration phase is used to determine system parameters, fitting parameters, etc.; the application phase uses the calibration results to directly calculate the target exposure time; the details are described below.

[0050] like Figures 1 to 3 As shown, the automatic exposure method of a differential absorption spectrometer proposed by the present invention comprises the following steps in the application stage:

[0051] Step 1: Set the exposure time, aperture value and fitting parameters of the visible camera based on the calibrated fast exposure differential imaging spectroscopy system, and set the exposure time range [T min , T cali], and also obtain the original exposure time-pixel value curve corresponding to the differential absorption spectrometer;

[0052] Step 2: Under the current light intensity, the hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visual camera are simultaneously obtained, and the differential absorption spectrometer is calculated based on the three-channel values ​​and the fitting parameters at the exposure time T cali The pixel value under the current light intensity is combined with the bias value corresponding to the 0-second exposure in the original exposure time-pixel value curve to obtain the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity;

[0053] Step 3: Substitute the target pixel value of the hyperspectral data into the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity to obtain the target exposure time T;

[0054] Step 4: Determine whether the target exposure time T is within the exposure time range [T min , T cali ] will be less than the minimum exposure time T min The target exposure time T is set to T = T min , will be greater than the maximum exposure time T cali The target exposure time T is set to T = T cali , to correct the target exposure time T to the exposure time range [T min , T cali ];

[0055] Step 5: Set the corrected target exposure time T as the exposure time of the differential absorption spectrometer and feed it back to the differential absorption spectrometer to obtain spectral data.

[0056] Through steps one to five, this embodiment proposes a fast automatic exposure imaging method and a dual imaging system suitable for a differential absorption spectrometer. The visible camera system is cheap, easy to obtain, and the overall cost is low. By priori drawing the exposure time-pixel value curve (i.e., T-DN curve) of the imaging spectrometer, combined with the calculated spectral imaging pixel value (DN value) under the current light intensity, the exposure time corresponding to the expected DN value can be directly obtained, and the calculation is efficient and fast. In addition, there is no need to use a differential absorption spectrometer for light measurement, which saves the light measurement cycle and makes full use of the data acquisition time window, which can greatly improve the data acquisition efficiency of the equipment, realize fast automatic exposure imaging, and extend the service life of the mechanical shutter.

[0057] The calibration stage is before the application stage. After the rapid exposure differential imaging spectroscopy system is calibrated, the aperture size, exposure time of the visible camera and other parameters are fixed. The processor feeds back the exposure time to the differential absorption spectrometer based on the hyperspectral data output by the differential absorption spectrometer and the color image data output by the visible camera to achieve automatic exposure of the differential absorption spectrometer.

[0058] During the calibration phase, the calibration process of the rapid exposure differential imaging spectroscopy system is as follows (a1) to (a8):

[0059] (a1) Determine the target light intensity range of the differential absorption spectrometer based on the actual application scenario [L min , L max ] and the original exposure time-pixel value curve of the differential absorption spectrometer under different illuminations, L min is the minimum light intensity value, L max is the maximum light intensity value;

[0060] The differential absorption spectrometer obtains both spatial and spectral information of the target area. Usually, the energy E received by the detector is linearly related to the exposure time of the differential absorption spectrometer. Therefore, the exposure time is linearly related to the pixel value (DN value), that is, the T-DN curve. For a given differential absorption spectrometer, its T-DN curve is determined, so the T-DN curve under different light intensity conditions can be obtained, which should be a fan-shaped structure, such as Figure 3 As shown in (a), L in (a) min , L max It represents the minimum and maximum light intensity values ​​corresponding to the imaging target in the actual application scenario, and Bais is the imaging bias of the differential absorption spectrometer.

[0061] (a2) According to the noise level of the differential absorption spectrometer imaging system, the noise area Noise Zone1 of the maximum output value of the differential absorption spectrometer is obtained, so as to set the pixel value boundary condition [DN min , D.N. max ] and exposure time range [T min , T cali ];

[0062] like Figure 3 In (a), Noise Zone 1 represents the noise limit set according to the noise level of the differential absorption spectrometer imaging system. That is, when the pixel value (DN value) of the imaging system enters Noise Zone 1, the value may be invalid due to the presence of noise. Therefore, the maximum effective DN value is set to be slightly smaller than Noise Zone 1, that is, DN max , the maximum effective pixel value DN max Below 50% of the noise zone Noise Zone1, that is, DN max =NoizeNone1 min -NoiseZone1 width 0.5, NoizeNone1 min 、NoiseZone1 widthThey are the lower boundary of the noise zone Noise Zone1 and the width of the noise zone; at the maximum light intensity L max The corresponding exposure time on the original exposure time-pixel value curve (i.e., T-DN curve) is set to T cali , which is the exposure time when acquiring data in the calibration phase; combined with the minimum light intensity value L min The original exposure time-pixel value curve under the exposure time T cali At the minimum light intensity L min The minimum valid pixel value DN corresponding to min ; [DN min , D.N. max ] is the pixel value boundary condition of the differential absorption spectrometer in the calibration stage.

[0063] According to the maximum light intensity condition L max The original exposure time-pixel value curve under the condition of min The corresponding time is set as the minimum exposure time T min ; [T min , T cali ] is the exposure time range of the differential absorption spectrometer during the calibration stage.

[0064] (a3) According to the light intensity range of the observed target [L min , L max ] Determine the exposure time-grayscale value relationship curve of the visible camera at different apertures, and set the effective grayscale value range of the visible camera imaging in combination with the noise area Noise Zone2 of the visible camera [Gr min , Gr max ];

[0065] Similarly, in the imaging system of a visual camera, the relationship between the grayscale value and the exposure time is also linear under fixed exposure time and fixed light intensity, that is, the exposure time-grayscale value relationship curve (T-Gr curve). Here, the grayscale value is calculated from the values ​​of the three RGB channels through formula (1):

[0066] Gr=c1·R+c2·G+c3·B; (1)

[0067] Wherein, Gr represents the grayscale value of the visible camera, R, G, and B are the three-channel values ​​of the visible camera, namely, the values ​​of the red, green, and blue channels, c1, c2, and c3 are independent parameters, preferably, c1=0.299, c2=0.587, and c3=0.114.

[0068] The same method as the imaging spectrometer can be used to obtain Figure 3 The figure in (b), that is: set the maximum effective gray value Gr maxSlightly smaller than the noise zone Noise Zone2 of the visible camera, that is, the maximum effective gray value Gr max Below 50% of the noise zone Noise Zone2 of the visible camera, that is, Gr max =NoizeNone2 min -NoiseZone2 width 0.5, NoizeNone2 min 、NoiseZone2 width are the lower boundary of the noise zone Noise Zone2 and the width of the noise zone, respectively, and the maximum effective grayscale value Gr under different apertures is obtained. max Corresponding to the maximum light intensity condition L max The maximum exposure time T s ; Combined with the minimum light intensity value L under different apertures min The corresponding exposure time-gray value relationship curve is obtained, and the exposure time T s At the minimum light intensity L min The minimum effective gray value Gr min .

[0069] Figure 3 (b) Gr max Represents the maximum effective grayscale value, which is slightly smaller than the noise zone Noise Zone2 and is obtained by the noise level of the visible camera imaging system and the maximum output grayscale value. Adjust the aperture size to aperture 1, aperture 2, ..., aperture N, and you can get the light intensity range [L min , L max ] under different exposure time-grayscale value regions (T-Gr region), similarly, the minimum effective grayscale value Gr can be obtained min ,[Gr min , Gr max ] is the effective grayscale value range of the visible camera imaging system.

[0070] (a4) Set the exposure time T2 of the visible camera to be less than the maximum exposure time T cali , the exposure time-grayscale value relationship curve under different apertures under the exposure time T2 is obtained, and the effective grayscale value range closest to the effective grayscale value range [Gr min , Gr max ] is used as the effective aperture, at which point the dynamic range of the visible camera is the largest;

[0071] In order to make full use of the dynamic range of the visible camera and realize the real-time output of the exposure time, the exposure time of the visible camera must be less than the exposure time T of the differential absorption spectrometer during the calibration stage. cali , if it is T2, then T2<T caliUnder T2 conditions, the exposure time-grayscale value relationship curve (T-Gr curve) under different aperture sizes can be obtained, and the effective grayscale value range closest to [Gr min , Gr max ], such as aperture 2. Figure 3 As shown in (c), the dynamic range of the visible camera is the largest at this time.

[0072] (a5) The exposure time and aperture size of the visual camera are set based on the exposure time T2 and the effective aperture.

[0073] (a6) based on the fact that the field of view of the visible camera covers the field of view of the differential absorption spectrometer, calibrating the field of view of the visible camera and the field of view of the differential absorption spectrometer;

[0074] The differential absorption spectrometer Rect(lt, lb, rt, tb) is marked in the visible camera field of view, where lt, lb, rt, tb represent the upper left, lower left, upper right and lower right coordinates of the rectangular local area of ​​the field of view, respectively.

[0075] (a7) Using the calibrated visible camera and differential absorption spectrometer, the target in the actual application scene is photographed to obtain the light intensity range of multiple observation targets [L min , L max ] The hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visual camera;

[0076] (a8) Based on the field of view in (a6), the common field of view data of the differential absorption spectrometer and the visible camera, the pixel value boundary conditions [DN min , D.N. max ] and the effective gray value range [Gr min , Gr max ], the three-channel numerical and hyperspectral data are fitted to obtain the relationship between the pixel values ​​of the differential absorption spectrometer and the pixel values ​​of the visible camera, and the fitting parameters are obtained, which are the calibration results.

[0077] Under the above conditions, the light intensity range [L min , L max ], and the grayscale value of the differential absorption spectrometer is obtained by fitting the pixel values ​​of the RGB channels of the visible camera, such as Figure 3 As shown in (d). The study found that under different lighting conditions, the spectral components of the imaging target will affect the fitting parameters, which are mainly related to the ultraviolet intensity. Therefore, according to the relationship between the three-channel bands of the visible camera, a piecewise function can be used for fitting:

[0078]

[0079] Among them, N is the pixel value of the differential absorption spectrometer, α, β, γ, α′, β′, and γ′ are all fitting parameters, R, G, and B are the three-channel values ​​of the visible camera, and δ and δ′ are balance parameters.

[0080] Therefore, in the practical application stage, the pixel value (DN value) of the differential absorption spectrometer can be obtained according to the pixel values ​​of the three channels R, G, and B obtained by the visual camera and combined with formula (2), which corresponds to the differential absorption spectrometer in step 2 at the exposure time T cali The pixel value under the light intensity can be combined with the corresponding bias Bias when the exposure time T = 0 to obtain the T-DN curve under the light intensity. Specifically, the bias Bias is taken as a point on the straight line and the exposure time T cali And the exposure time T cali The point corresponding to the pixel value under the light intensity is taken as another point, and a straight line is drawn. The straight line is the T-DN curve under the light intensity.

[0081] The target pixel value (DN) is calculated on the T-DN curve. tar The exposure time T under the value) is calculated, and the exposure time T is corrected based on step four, and the corrected exposure time T is fed back to the differential absorption spectrometer as the target exposure time T. The differential absorption spectrometer can obtain high signal-to-noise ratio hyperspectral data according to the target exposure time T fed back by the processor.

[0082] This embodiment proposes a method for automatic exposure calibration of a differential imaging spectrometer. Combined with actual application scenarios, the three-channel data of the differential absorption spectrometer and the visible camera RGB are obtained, and fitting parameters are obtained by a piecewise function fitting method. The fitting parameters are obtained through system calibration, and a method for drawing an exposure time-pixel value curve (T-DN curve) and an exposure time-gray value relationship curve (T-Gr curve) is proposed in combination with the working principles of the imaging spectrometer and the visible camera imaging system. The dynamic range of the two imaging systems can be fully utilized to improve the calibration accuracy of the system.

[0083] In addition, the frame rate of the visible camera provided in this embodiment is higher than that of the imaging spectrometer, which can realize real-time adjustment of the exposure time parameters of the differential absorption spectrometer and improve the data acquisition efficiency of the observation time window; and there is no need for multiple exposure imaging, which reduces the number of imaging times of the differential absorption spectrometer, and can greatly increase the shutter service life of the mechanical shutter imaging spectrometer.

[0084] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An automatic exposure method for a differential absorption spectrometer, characterized in that: The steps include: Step 1: Set the exposure time, aperture value and fitting parameters of the visible camera based on the calibrated fast exposure differential imaging spectroscopy system, and set the exposure time range [T min ,T cali ], and also obtain the original exposure time-pixel value curve corresponding to the differential absorption spectrometer; Step 2: Under the current light intensity, the hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visual camera are simultaneously obtained, and the differential absorption spectrometer is calculated based on the three-channel values ​​and the fitting parameters at the exposure time T cali The pixel value under the current light intensity is combined with the bias value corresponding to the 0-second exposure in the original exposure time-pixel value curve to obtain the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity; Step 3: Substitute the target pixel value of the hyperspectral data into the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity to obtain the target exposure time T; Step 4: Determine whether the target exposure time T is within the exposure time range [T min ,T cali ] will be less than the minimum exposure time T min The target exposure time T is set to T = T min , will be greater than the maximum exposure time T cali The target exposure time T is set to T = T cali , to correct the target exposure time T to the exposure time range [T min ,T cali ]; Step 5: Set the corrected target exposure time T as the exposure time of the differential absorption spectrometer and feed it back to the differential absorption spectrometer to obtain spectral data.

2. The automatic exposure method of the differential absorption spectrometer according to claim 1, characterized in that: The calibration process of the rapid exposure differential imaging spectroscopy system is as follows: (a1) Determine the target light intensity range of the differential absorption spectrometer based on the actual application scenario [L min ,L max ] and the original exposure time-pixel value curve of the differential absorption spectrometer under different illumination, L min is the minimum light intensity value, L max is the maximum light intensity value; (a2) According to the noise level of the differential absorption spectrometer imaging, the noise area Noise Zone1 of the maximum output value of the differential absorption spectrometer is obtained, so as to set the pixel value boundary condition [DN min ,DN max ] and exposure time range [T min ,T cali ]; (a3) According to the light intensity range of the observed target [L min ,L max ] Determine the exposure time-grayscale value relationship curve of the visible camera at different apertures, and set the effective grayscale value range of the visible camera imaging in combination with the noise area Noise Zone2 of the visible camera [Gr min ,Gr max ]; (a4) Set the exposure time T2 of the visible camera to be less than the maximum exposure time T cali , the exposure time-grayscale value relationship curve under different apertures under the exposure time T2 is obtained, and the effective grayscale value range closest to the effective grayscale value range [Gr min ,Gr max ] is used as the effective aperture, at which point the dynamic range of the visible camera is the largest; (a5) setting the exposure time and aperture size of the visual camera based on the exposure time T2 and the effective aperture; (a6) based on the fact that the field of view of the visible camera covers the field of view of the differential absorption spectrometer, calibrating the field of view of the visible camera and the field of view of the differential absorption spectrometer; (a7) Using the calibrated visible camera and differential absorption spectrometer, the target in the actual application scene is photographed to obtain the light intensity range of multiple observation targets [L min ,L max ] The hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visual camera; (a8) Based on the field of view in (a6), the common field of view data of the differential absorption spectrometer and the visible camera, the pixel value boundary conditions [DN min ,DN max ] and the effective gray value range [Gr min ,Gr max ], the three-channel numerical and hyperspectral data are fitted to obtain the relationship between the pixel values ​​of the differential absorption spectrometer and the pixel values ​​of the visible camera, and the fitting parameters are obtained, which are the calibration results.

3. The automatic exposure method of the differential absorption spectrometer according to claim 2, characterized in that: In (a2), the pixel value boundary condition of the differential absorption spectrometer during the calibration phase [DN min ,DN max ] and exposure time range [T min ,T cali The setting process of ] is as follows: Set the maximum effective pixel value DN max Below 50% of the noise area Noise Zone1, the maximum effective pixel value DN is obtained max Corresponding to the maximum light intensity condition L max The maximum exposure time T cali ; Combined with the minimum light intensity value L min The original exposure time-pixel value curve under the exposure time T cali At the minimum light intensity l min The minimum valid pixel value DN corresponding to min ; According to the maximum light intensity condition L max The original exposure time-pixel value curve under the condition of min The corresponding time is set as the minimum exposure time T min .

4. The automatic exposure method of the differential absorption spectrometer according to claim 2, characterized in that: In (a3), the effective grayscale value range of the visible camera imaging [Gr min ,Gr max The setting process of ] is as follows: Set the maximum effective grayscale value Gr max The maximum effective gray value Gr under different apertures is obtained by setting it to 50% of the noise area Noise Zone2 of the visible camera. max Corresponding to the maximum light intensity condition L max The maximum exposure time T s ; Combined with the minimum light intensity value L at different apertures min The corresponding exposure time-gray value relationship curve is obtained, and the exposure time T s At the minimum light intensity L min The minimum effective gray value Gr min .

5. The automatic exposure method of the differential absorption spectrometer according to claim 2, characterized in that: In (a8): the three-channel numerical value and hyperspectral data are fitted to obtain the relationship between the pixel value of the differential absorption spectrometer and the pixel value of the visible camera, and the fitting parameters are obtained, where the relationship formula between the pixel value of the differential absorption spectrometer and the pixel value of the visible camera is as follows: Among them, N is the pixel value of the differential absorption spectrometer, α, β, γ, α′, β′, and γ′ are all fitting parameters, R, G, and B are the three-channel values ​​of the visible camera, and δ and δ′ are balance parameters.

6. The automatic exposure method of the differential absorption spectrometer according to claim 5, characterized in that: In step 2, based on the three-channel values ​​and the fitting parameters and using the relationship formula between the pixel value of the differential absorption spectrometer and the pixel value of the visible camera, the differential absorption spectrometer at the exposure time T is calculated. cali The pixel value below.

7. An automatic exposure system for a differential absorption spectrometer, characterized in that: Implementing automatic exposure of a differential absorption spectrometer based on a calibrated rapid exposure differential imaging spectrometer system, wherein the rapid exposure differential imaging spectrometer system comprises a differential absorption spectrometer, a visual camera, and a processor; The calibrated rapid exposure differential imaging spectroscopy system specifically performs the following process: Step 1: Set the exposure time, aperture value and fitting parameters of the visible camera based on the calibrated fast exposure differential imaging spectroscopy system, and set the exposure time range [T min ,T cali ], and also obtain the original exposure time-pixel value curve corresponding to the differential absorption spectrometer; Step 2: Under the current light intensity, the processor simultaneously obtains the hyperspectral data output by the differential absorption spectrometer and the three-channel values ​​output by the visible camera, and calculates the differential absorption spectrometer at the exposure time T based on the three-channel values ​​and the fitting parameters. cali The pixel value under the current light intensity is combined with the bias value corresponding to the 0-second exposure in the original exposure time-pixel value curve to obtain the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity; Step 3: The processor brings the target pixel value of the hyperspectral data into the exposure time-pixel value curve of the differential absorption spectrometer under the current light intensity to obtain the target exposure time T; Step 4: The processor determines whether the target exposure time T is within the exposure time range [T min ,T cali ] will be less than the minimum exposure time T min The target exposure time T is set to T = T min , will be greater than the maximum exposure time T cali The target exposure time T is set to T = T cali , to correct the target exposure time T to the exposure time range [T min ,T cali ]; Step 5: The processor sets the corrected target exposure time T as the exposure time of the differential absorption spectrometer and feeds it back to the differential absorption spectrometer to obtain spectral data.

Images