Measurement range expansion method of spectral light field imaging temperature measurement system and imaging device

Abstract

The invention discloses a measurement range expansion method of a spectral light field imaging temperature measurement system and an imaging device. The measurement range expansion method of the spectral light field imaging temperature measurement system comprises the following steps: establishing a response model of a temperature measurement system spectral channel through a blackbody furnace calibration experiment; and establishing a target function, and optimizing the established target function to obtain the attenuation slice transmittance of the central wavelength of the channel. The measurement range of the spectral light field imaging temperature measurement system is expanded by obtaining the transmittance of the attenuation sheet with the central wavelength. The method provided by the invention can be suitable for temperature measurement objects with different spectral characteristics, and ensures the consistency of dynamic response of each channel and broadens the measurement range of a temperature measurement system by combining a specified temperature measurement interval and adjusting the transmittance parameter of the attenuation slice.

Description

Technical Field

[0001] The present invention relates to the field of spectral light field imaging and radiation temperature measurement, and in particular to a method for extending the measurement range of a spectral light field imaging temperature measurement system and an imaging device. Background Art

[0002] Incomplete combustion of hydrocarbon fuels not only reduces energy efficiency, but also poses a serious threat to global warming and human health. Accurate temperature measurement not only helps to deepen the research on the mechanism of soot particle formation, but also provides a measurement basis for pollutant control in industrial combustion devices. The measurement method based on light field imaging technology can obtain the transmission process of light in three-dimensional space, and reconstruct the distribution of physical parameters by establishing a coupling model between soot optical parameters, temperature and detection light intensity. In actual processes, soot optical parameters are often unknown. Therefore, this technology requires coupling spectral information to solve physical parameters and temperature. Generally, the spectral information of three fixed bands of Red (R), Green (G) and Blue (B) in the visible light band is obtained by coupling Bayer filters at the detector pixels. However, in this way, the spectral channel is too wide, which easily leads to an increase in the error of the detection light intensity value, thereby affecting the reconstruction result. The increase in the number of spectral channels of the sensor filter will cause serious aliasing effects, and the current production cost is relatively high due to the limitation of processing technology.

[0003] The temperature measurement system based on spectral light field imaging can effectively solve the problems of wide spectral channels and information redundancy in light field imaging. It has the advantages of flexible spectral channel selection, narrow channel range, clear spatial-spectral modulation relationship, and simple reconstruction algorithm. The implementation method is to insert a filter array in the aperture plane of the main lens to convert redundant light field information into spectral information. Therefore, the acquisition of spectral light field information can be completed through a single exposure. However, in the temperature measurement process, the temperature measurement system has a prominent problem of narrow channel response range. The main reason is that it is limited by the spectral characteristics of the object to be measured and the spectral dynamic response of the sensor itself in the system, resulting in excessive differences in the dynamic response between radiation energy and grayscale under different channels. After the gray of the high-response channel is overexposed, the signal crosstalk generated will affect the channel with low response characteristics, thereby reducing the effective response range of the overall system, making it difficult for the spectral light field imaging temperature measurement system to cover the target temperature range of the measured object, and the measurement range becomes narrower. Summary of the invention

[0004] In view of the shortcoming that the measurement range of the current spectral light field imaging temperature measurement system is too narrow, the present invention provides a measurement range expansion method and an imaging device of the spectral light field imaging temperature measurement system.

[0005] The solution of the present invention to solve the technical problem is:

[0006] The present invention first provides a method for extending the measurement range of a spectral light field imaging temperature measurement system, comprising:

[0007] Using a fiber optic spectrometer to collect the spectral characteristics of the target object to be measured, and obtaining the emission peak wavelength of the target object to be measured during the combustion process;

[0008] Within the response wavelength range of the temperature measurement system, avoid the emission peak band generated by the combustion process of the target object to be measured, and arbitrarily select the central wavelength λ of each channel under the condition of ensuring that the central wavelength spacing is 20-30nm. The number of channels is Channel.

[0009] The channel response model of the spectral light field imaging temperature measurement system is established through the blackbody furnace calibration experiment:

[0010]

[0011] Where h1 and h2 are Planck constants, which are 3.7148×10 -16 / (W·m 2 ),1.4388×10 -2 / (m·K); gray λ is the grayscale response of the sensor under the corresponding spectral channel; Δt is the exposure time of the imaging device, in ms; T λ,min It is the black body furnace temperature corresponding to the grayscale response value of 2%FR under the exposure parameters during the shooting of the object to be tested, and the unit is K; T λ,max It is the temperature of the black body furnace when the grayscale response value of the current channel is 90% FR, and the unit is K; FR represents the full scale; Φ λ is the transmittance of the attenuator with a central wavelength of λ; Δλ is the half-peak width of the spectral channel, in nm; E λ is the spectral radiation intensity of the channel center wavelength λ, in W / (m 3 sr); is the coefficient of the corresponding relationship under the specified channel, in J·s p-1 / (m 2 ·sr); b λ,Noise is the constant term of the relationship, and its unit is W / (m 2 ·sr); q is the highest degree of the relationship, q≤4; M is the number of calibration points in the blackbody furnace experiment;

[0012] Establish the objective function F:

[0013]

[0014] Where N is the number of particle dimensions, and its value is the number of channels excluding the reference channel, that is, N = Channel-1; is the theoretical value of the mth calibration point under the nth channel; It is the theoretical value of the mth calibration point under the reference channel with the lowest response characteristic. The attenuation rate of the channel with low response characteristic defaults to 100%. ; M is the number of theoretical grayscale calculation points of the channel;

[0015] Optimize the established objective function to obtain the transmittance Φ of the attenuator with a central wavelength of λ λ .

[0016] Optimize the established objective function to obtain the transmittance Φ of the attenuator with a central wavelength of λ λ In the step, the particle swarm optimization algorithm is used to optimize the established objective function.

[0017] The method of optimizing the established objective function using the particle swarm optimization algorithm includes:

[0018] Set the number of particle swarms Num, particle dimension N, and maximum particle position x max =1, minimum value of particle position x min =0, maximum particle velocity v max =0.005, minimum particle velocity v min =0.0001, maximum iteration step length Loop max =1000, precision iteration termination threshold ε1 = 10 -3 and step size iteration termination threshold ε2 = 10 -4 ;

[0019] Randomly initialize the particle position X = [x 1, ...,x i ,...,x Num ], the dimension is N×Num; x i is the current iteration position of the particle in the ith particle group, with a dimension of N×1; the speed V = [v 1, ...,v i ,...,v Num ], dimension is N×Num; v i is the current velocity of the particle in the ith particle group, with a dimension of N×1. Set the optimal position of the individual particle group according to the initial positions of all particles; P = [p 1, ...,p i ,...,p Num ], with dimension N×Num and global optimal position G=[g 1, ...,g j ,...,g N ] Τ , dimension is N×1; p Num It is the individual optimal position of the Numth particle group, with a dimension of N×1.

[0020] x i =x min ·E+(x max -x min )·r x

[0021] v i =v min ·E+(v max -v min )·r v

[0022] G=p1

[0023] In the formula, r x It is a random number for initial assignment of particle position, with the element value interval of [0,1] and dimension of N×1; r v It is a random number for initial assignment of particle velocity, with an element value interval of [0,1] and a dimension of N×1; E is a vector of all 1s and a dimension of N×1;

[0024] Update the position of the particle, calculate the objective function value F corresponding to the particle, and update the individual optimal and global optimal values:

[0025] v i =ω×v i +c1×r1×(p i -x i )+c2×r2×(g i -x i )

[0026] x i =x i +v i

[0027] p i =x i if F(x i ) <F(p i )

[0028] G=p i if F(p i ) <F(G)

[0029] Where ω is the inertia weight, which is selected as 0.65; c1 and c2 are acceleration constants, which are set to 1.48, indicating the degree of convergence of particles to the individual optimal and global optimal positions; r1 and r2 are diagonal matrices composed of updated random numbers, with the interval of elements being [0,1].

[0030] The position of each particle in the updated particle swarm is checked to prevent the position from exceeding the boundary set by the algorithm; if it exceeds, the position of each particle is trimmed to within the boundary:

[0031] x i =max(x min ×E,min(x i ,x max ×E))

[0032] Determine whether the current number of loops has reached the maximum number of iterations, or whether the global optimal value is lower than the set threshold ε1 (i.e., F(G)≤ε1), or whether the reduction in the global optimal value is lower than the set threshold ε2 (i.e., ΔF(G)|≤ε2); if one of these three judgment conditions is met, then exit the loop, otherwise increase the current number of loops by one, re-update the positions of each particle, and enter the next loop;

[0033] The global optimal solution vector G obtained by the algorithm is matched one by one with the transmittance Φ of the corresponding spectral channel. The dimension is N×1, where the channel center wavelength is sorted from small to large, corresponding to the transmittance, that is, λ1<...<λ j <...<λ N , and its corresponding formula is:

[0034] Set the number of particle swarm Num = 300, particle dimension N = channel-1; the maximum particle position x max =1,x min =0, v max =0.005,v min =0.0001, maximum iteration step length Loop max =1000, iteration termination threshold ε1 = 10 -3 and ε2=10 -4 .

[0035] The inertia weight ω is selected as 0.65.

[0036] The acceleration constants c1 and c2 were set to 1.48.

[0037] The present invention also provides an imaging device corresponding to the spectral light field imaging temperature measurement system, including a main lens, a microlens array and a sensor. The main lens and the microlens array jointly image the object on the sensor. A filter attenuation plate array for expanding the temperature measurement range is arranged at the aperture of the main lens. The array elements of the filter attenuation plate array have a filtering part for transmitting different central wavelengths and an attenuation part for attenuating different central wavelengths.

[0038] The transmittance of the attenuation part of the filter attenuation plate array is determined based on the measurement range expansion method of the spectral light field imaging temperature measurement system described in any one of claims 1-6.

[0039] The array elements of the filter attenuation plate array are two layers of optical elements, and the two layers of optical elements are filters and attenuation plates respectively.

[0040] Compared with the prior art, the present invention has the following advantages:

[0041] The present invention reduces the dynamic response difference between channels by setting filter attenuation plates with different transmittances in different channels, thereby broadening the temperature measurement range of the system, so that the spectral light field imaging temperature measurement system can cover the target temperature range of the measurement object.

[0042] The present invention proposes a two-layer optical element that bonds the attenuation plate and the filter plate, which broadens the temperature measurement range of the system. By replacing the attenuation plate array with different transmittances, the measurement requirements of the measurement object in different temperature ranges can be met, with low cost and simple operation.

[0043] The present invention is applicable to target objects with different spectral characteristics, can avoid the emission peak band of the object to be measured, arbitrarily select the central wavelength of the channel within the response wavelength range of the temperature measurement system, and obtain high signal-to-noise ratio grayscale information of multiple spectral channels through a single exposure.

[0044] The present invention proposes to use a particle swarm optimization (PSO) algorithm with strong global search capability to optimize the transmittance parameters of the spectral channel, which can overcome the problem that traditional algorithms are prone to falling into local extreme values. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of the imaging device structure of the spectral light field imaging temperature measurement system;

[0046] Figure 2 This is a schematic diagram of the blackbody furnace temperature measurement experimental device of the spectral light field imaging temperature measurement system;

[0047] Figure 3 Algorithm flow chart for calculating transmittance parameters for particle swarm optimization (PSO) algorithm;

[0048] Figure 4 It is the iterative result of the particle swarm optimization (PSO) algorithm for transmittance parameters under different channels;

[0049] Figure 5 This is a schematic diagram of gluing the attenuation plate and the filter array;

[0050] Figure 6 This is a comparison of the before and after temperature measurement range expansion of the four-wavelength temperature measurement system with a blackbody furnace as the temperature measurement object. DETAILED DESCRIPTION

[0051] The present invention is further described below in conjunction with the accompanying drawings and specific examples. It should be understood that these examples are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, modifications of various equivalent forms of the present invention by those skilled in the art all fall within the scope defined by the appended claims of this application.

[0052] Example 1

[0053] This embodiment uses a black body furnace as the temperature measurement object, and a four-channel spectral light field imaging temperature measurement system as the target for widening the dynamic range. The temperature measurement range is 1583-2300K, and the emissivity is 0.99. The detailed steps are as follows:

[0054] Step 1: Since the temperature measurement object is a blackbody furnace, the wavelength range of the spectral channel to be tested is set within the visible light range, and the temperature measurement system equipped with only the filter array is subjected to a blackbody furnace calibration test. Figure 2 As shown, during the shooting process, the distance between the system and the blackbody opening of the blackbody furnace is 70cm, and the exposure parameters are set to 0.1, 0.5, 1, 2, and 4ms respectively. The grayscale response of the system is recorded ten times every 50K and the average value is taken. The minimum temperature of the blackbody furnace is set to 1583K, and the maximum temperature is set to 2300K. The temperature measurement system sets 4 different channels (Channel=4), and the central wavelengths are 540, 610, 640, and 660nm, respectively, and the corresponding half-peak widths are 40, 35, 25, and 25nm. Through polynomial fitting, the corresponding response formula of the system under four channels can be obtained, and the results are shown in Table 1.

[0055] Table 1. Fitting results of four spectral channels of blackbody furnace calibration experimental system

[0056]

[0057] Step 2: Since the temperature measurement object is set to a black body furnace, for the four different channels, the corresponding minimum calibration temperature and maximum calibration temperature are the same. The center wavelength of the low response channel is 540nm, and the corresponding transmittance is 100%. 540nm =100%. That is:

[0058] T 540,min =T 610,min =T 640,min =T 660,min =1583K

[0059] T 540,max =T 610,max =T 640,max =T 660,max =2300K

[0060] Step 3: Reference Figure 3 The algorithm flow chart uses the particle swarm optimization (PSO) algorithm to determine the appropriate transmittance parameters under different channels. In the temperature range of step 2, a verification point is selected every 25K, with a total of 14 verification points (M=14). The theoretical calculated value of the 540nm channel is used as a reference value and the objective function F is established, as shown in formula (1). Set the number of particle swarms to Num=300, the particle dimension N=3, and the maximum particle position x max =1,x min =0,v max =0.005,v min =0.0001, maximum iteration step length Loop max =1000, the iteration termination threshold is ε1=10 -3 and ε2=10 -4 According to formula (2), the position of the particle is randomly initialized X = [x 1, …,x i ,...,x 300 ], the dimension is 3×300, and the velocity V = [v 1, …,v i ,...,v 300 ], the dimension is 3×300, and the individual optimal position of the particle is set to P = [p 1, …,p i ,...,p 300 ], the dimension is 3×300 and the global optimal position G = [g 1, …,g j ,...,g3], the dimension is 3×1.

[0061]

[0062]

[0063] In the formula, x i is the current iteration position of the particle in the i-th particle group, with a dimension of 3×1; v i is the current velocity of the particle in the i-th particle group, with a dimension of 3×1; r x It is a random number for initial assignment of particle position, the value range of the element is [0,1], and the dimension is 3×1; r v It is a random number for initial assignment of particle velocity. The value range of the elements is [0,1] and the dimension is 3×1. E is an all-1 vector with a dimension of 3×1.

[0064] Step 4: Update the position of the particle according to formula (3), calculate the objective function value F corresponding to the particle according to formula (1) and the response model relationship in the invention content, and update the individual optimal and global optimal values ​​according to formula (4).

[0065]

[0066]

[0067] In the formula, ω is the inertia weight selected as 0.65. c1 and c2 are acceleration constants, set to 1.48, indicating the degree of convergence of particles to the individual optimal and global optimal positions; r1 and r2 are diagonal matrices composed of updated random numbers, with the element interval being [0,1].

[0068] The position of each particle in the updated particle swarm is checked to prevent it from exceeding the boundary set by the algorithm. If it exceeds the boundary, it is trimmed to within the boundary:

[0069] x i =max(x min ×E,min(x i ,x max ×E)) (5)

[0070] Determine whether the current number of loops has reached the maximum number of iterations, or whether the global optimal value is lower than the set threshold ε1 (i.e., F(G)≤ε1), or whether the reduction in the global optimal value is lower than the set threshold ε2 (i.e., ΔF(G)|≤ε2). If one of these three judgment conditions is met, then the loop is exited, otherwise the current number of loops is increased by one (i.e., Loop=Loop+1), and the process jumps to step 4 to re-update the positions of each particle and enter the next loop.

[0071] Step 5: The global optimal solution vector G obtained by the algorithm and the transmittance Φ of the corresponding spectral channel are matched one by one. The dimension is N×1, where the channel center wavelength is sorted from small to large, corresponding to the transmittance, that is, λ1<...<λ j <...<λ3,λ1=610nm,λ2=640nm,λ3=660nm, the corresponding formula is: The solution of its transmittance changes with the iteration step size as follows Figure 4 shown.

[0072] The transmittance parameters of the four channels obtained by algorithm calculation are Φ 540nm =100%,Φ 610nm =22.38%,Φ 640nm =24.38%,Φ 660nm =10.53%. Considering the cost, the transmittance parameter of the attenuator array can be selected close to the four theoretical values. Therefore, the actual transmittance of the four-wavelength attenuator array is: Φ 540nm =100%,Φ 610nm =20%,Φ640nm =25%,Φ 660nm = 10%. The imaging device structure corresponding to the system is as follows Figure 1 As shown, the attenuation plate array and the filter array are connected according to Figure 5 The calibrated spectral light field imaging temperature measurement system was built, and the black body furnace experiment was conducted again to calibrate the channel dynamic response curve. The temperature measurement exposure time was 0.4ms. The corrected dynamic response curve is shown in the figure below. Figure 6 The maximum grayscale deviation of the method proposed in the present invention at four channels of 540nm, 610nm, 640nm, and 660nm is 3.4, the minimum deviation is 0.8, and the average deviation is 1.9. The data points are fitted with a curve to obtain the correlation coefficient R 2 is 0.999, which means that all points are on the same line, that is, the dynamic responses of each channel are basically the same. In addition, the length of the temperature measurement interval is increased from 228K to 612K, which is 2.7 times higher. This verifies the feasibility of this method in broadening the measurement range of the spectral light field imaging temperature measurement system in practical applications.

[0073] Example 2

[0074] This embodiment is an imaging device of a spectral light field imaging temperature measurement system. Figure 1 The imaging device of the spectral light field temperature measurement system includes a main lens, a filter attenuation sheet array, a microlens array and a sensor. The main lens and the microlens array jointly image the object on the sensor; the filter attenuation sheet array is arranged at the aperture of the main lens to expand the temperature measurement range, and the array element of the filter attenuation sheet array has a filtering part for transmitting different central wavelengths and an attenuation part for attenuating different central wavelengths.

[0075] The transmittance of the attenuation part of the filter attenuation plate array is determined based on the measurement range expansion method of the spectral light field imaging temperature measurement system provided in Example 1.

[0076] In one embodiment, the array element of the filter attenuation plate array is a layer of optical elements made of a material with a specific transmittance and a central wavelength.

[0077] In one embodiment, the array elements of the filter attenuator array are two layers of optical elements, which are filters and attenuators respectively. The filters are used to select the central wavelength, and the attenuators have a specific transmittance.

Claims

1. A method for extending the measurement range of a spectral light field imaging temperature measurement system, comprising: Using a fiber optic spectrometer to collect the spectral characteristics of the target object to be measured, and obtaining the emission peak band of the combustion process of the target object to be measured; Within the response wavelength range of the temperature measurement system, avoid the emission peak wavelength of the combustion process of the target object to be measured, and arbitrarily select the central wavelength λ under the channel under the condition of ensuring that the channel wavelength spacing is 20-30nm, and determine the number of channels Channel; The channel response model of the spectral light field imaging temperature measurement system is established through the blackbody furnace calibration experiment: Where h1 and h2 are Planck constants, which are 3.7148×10 -16 / (W·m 2 ),1.4388×10 -2 / (m·K); gray λ is the grayscale response of the sensor under the corresponding spectral channel; Δt is the exposure time of the imaging device; T λ,min It is the black body furnace temperature corresponding to the grayscale response value of 2%FR under the exposure parameters during the shooting of the object to be tested; T λ,max It is the temperature of the black body furnace when the grayscale response value of the current channel is 90% FR; FR represents the full scale; Φ λ is the transmittance of the attenuator of the channel with a central wavelength of λ; Δλ is the half-peak width of the spectral channel; E λ is the spectral radiation intensity of the channel center wavelength λ; is the coefficient of the corresponding relationship under the specified channel; b λ,Noise is the constant term of the relation; q is the highest degree of the relation, q≤4; Establish the objective function F: Where N is the number of particle dimensions, N = Channel-1, Channel is the number of channels; is the theoretical value of the mth calibration point under the nth channel; It is the theoretical value of the mth calibration point under the reference channel with the lowest response characteristic, and the attenuation rate of the channel with low response characteristic defaults to 100%; M is the number of grayscale theoretical calculation points of the channel; Optimize the established objective function to obtain the transmittance Φ of the attenuator with a central wavelength of λ λ .

2. The method for extending the measurement range of a spectral light field imaging temperature measurement system according to claim 1, characterized in that: Optimize the established objective function to obtain the transmittance Φ of the attenuator with a central wavelength of λ λ In the step, the particle swarm optimization algorithm is used to optimize the established objective function.

3. The method for extending the measurement range of a spectral light field imaging temperature measurement system according to claim 2, characterized in that: The method of optimizing the established objective function using the particle swarm optimization algorithm includes: Set the number of particle swarms Num, particle dimension N, and maximum particle position x max =1, minimum value of particle position x min =0, maximum particle velocity v max =0.005, minimum particle velocity v min =0.0001, maximum iteration step length Loop max =1000, precision iteration termination threshold ε1 = 10 -3 and step size iteration termination threshold ε2 = 10 -4 ; Randomly initialize the positions of all particles X = [x1,…,x i ,...,x Num ], the dimension is N×Num; x i is the current iteration position of the particle in the ith particle group, with a dimension of N×1; the velocity V=[v1,…,v i ,...,v Num ], dimension is N×Num; v i is the current velocity of the particle in the i-th particle group, with dimension N×1; According to the initialization positions of all particles, the optimal position of the particle swarm is set to P = [p1,…,p i ,...,p Num ], with dimension N×Num and global optimal position G=[g1,…,g j ,...,g N ] Τ , dimension is N×1; p Num is the individual optimal position of the Num-th particle group, with a dimension of N×1; x i =x min ·E+(x max -x min )·r x v i =v min ·E+(v max -v min )·r v G=p1 In the formula, r x It is a random number for initial assignment of particle position, the value range of the element is [0,1], and the dimension is N×1; r v It is a random number for initial assignment of particle velocity, the value range of the element is [0,1], and the dimension is N×1; E is a vector of all 1s, and the dimension is N×1; Update the position of the particle, calculate the objective function value F corresponding to the particle, and update the individual optimal and global optimal values: v i =ω×v i +c1×r1×(p i -x i )+c2×r2×(g i -x i ) x i =x i +v i p i =x i if F(x i )<F(p i ) G=p i if F(p i )<F(G) Where ω is the inertia weight; c1 and c2 are acceleration constants; r1 and r2 are diagonal matrices composed of updated random numbers, with the element interval being [0,1]. The position of each particle in the updated particle swarm is checked to prevent the position from exceeding the boundary set by the algorithm; if it exceeds, the position of each particle is trimmed to within the boundary: x i =max(x min ×E,min(x i ,x max ×E)) Determine whether the current number of loops has reached the maximum number of iterations, or whether the global optimal value is lower than the set threshold ε1, F(G)≤ε1, or the reduction of the global optimal value is lower than the set threshold ε2, ΔF(G)|≤ε2; if one of the three judgment conditions is met, then jump out of the loop, otherwise the current number of loops is increased by one, and the positions of each particle are updated again to enter the next loop; The global optimal solution vector G obtained by the algorithm is matched with the transmittance Φ of the corresponding spectral channel one by one; The dimension is N×1, where the channel center wavelength is sorted from small to large, corresponding to the transmittance, that is, λ1<...<λ j <...<λ N , and its corresponding formula is:

4. The method for extending the measurement range of a spectral light field temperature measurement system according to claim 3, characterized in that: Set the number of particle swarms Num = 300, the particle dimension N = Channel-1; the maximum value of particle position x max =1,x min =0,v max =0.005,v min =0.0001, maximum iteration step length Loop max =1000, iteration termination threshold ε1 = 10 -3 and ε2=10 -4 .

5. The method for extending the measurement range of a spectral light field imaging temperature measurement system according to claim 3, characterized in that: The inertia weight ω is selected as 0.

65.

6. The method for extending the measurement range of a spectral light field imaging temperature measurement system according to claim 3, characterized in that: The acceleration constants c1 and c2 were set to 1.

48.

7. An imaging device of a spectral light field imaging temperature measurement system, comprising a main lens, a microlens array and a sensor, wherein the main lens and the microlens array jointly image an object on the sensor, characterized in that: A filter attenuation plate array for expanding the temperature measurement range is arranged at the aperture of the main lens, and an array element of the filter attenuation plate array has a filtering part for transmitting different central wavelengths and an attenuation part for attenuating different central wavelengths.

8. The imaging device according to claim 1, characterized in that In the imaging device of the system, the transmittance of the attenuation part of the filter attenuation plate array is determined based on the measurement range expansion method of the spectral light field imaging temperature measurement system described in any one of claims 1-6.

9. The imaging device according to claim 8, characterized in that: The array element of the filter attenuation plate array is a layer of optical elements.

10. The imaging device according to claim 8, characterized in that The array elements of the filter attenuation plate array are two layers of optical elements, and the two layers of optical elements are filters and attenuation plates respectively.

Images