A method for optimizing the design of spectral filtering channels based on target features

Abstract

The present invention relates to a method for optimizing the design of a spectral filtering channel based on target features. For each layer of thin film in the filtering channel, the present invention generates m candidate solutions, combines a genetic algorithm to perform crossover and mutation on the candidate solutions, expands the candidate solution set, eliminates the need to repeatedly debug the weights in the optimization algorithm, simplifies the computational complexity of the optimization algorithm, and then constructs an objective function including a penalty term based on considering the diversity of target feature peaks to obtain the film system combination of the optimal filtering channel from the mutated candidate solution set, realizing the free regulation of the central wavelength and bandwidth of the filtering channel, while preventing the coupling and overlap between multiple feature peaks of the same target, effectively overcoming the limitations of existing optimization algorithms, and making the design result more targeted.

Description

Technical Field

[0001] The present invention relates to the technical fields of spectral imaging and optical thin films. Specifically, it relates to a method for optimizing the design of spectral filtering channels based on target characteristics. Background Art

[0002] In the field of spectral detection, accurately identifying the spectral characteristics of targets is crucial for key application fields such as remote sensing detection and biomedical diagnosis. The filtering channel, as the core component in the spectral detection system, realizes the selective transmission of light at specific wavelengths while effectively suppressing out-of-band radiation. The filtering channel plays a crucial role in distinguishing the spectral characteristics of different targets. Specifically, the spectral characteristic peaks of different targets are different, and complex targets may exhibit multiple characteristic peaks. Therefore, in order to improve the accuracy and reliability of target detection, the key lies in flexibly adjusting the central wavelength and bandwidth of the filtering channel to precisely match the characteristic spectral peaks of different targets.

[0003] The filtering channel is usually composed of multiple thin films with different refractive indexes, which produce differences in the spatial and spectral distributions of light through the thin film interference effect. The existing optimization design methods for filtering channels are mainly divided into two categories: the analytical method and the optimization method. The analytical method relies on a specific film system structure and combines analytical design methods such as the equivalent layer method, the vector diagram method, or the equivalent interface method to manually design and adjust the film system. This method is time-consuming and laborious, and has high requirements for the designer's basic knowledge of thin film optics and film system design experience, etc. The optimization method is based on various optimization algorithms. By formulating an objective function, it uses numerical simulation and iterative search to minimize the difference between the design and the target spectrum and find the optimal structure. However, the existing optimization methods have limitations: i) The parameter scanning or gradient information-based optimization algorithms are prone to falling into local optimal solutions; ii) The existing objective function only includes two index values, the rectangularity and the ripple coefficient. Since the orders of magnitude of the rectangularity and the ripple coefficient are different, the weight values need to be repeatedly modified and debugged, and the weight values have a great influence on the accuracy of the optimization algorithm; iii) The problems such as processing errors are not fully considered. Summary of the Invention

[0004] The technical problem to be solved by the present invention is how to break through the limitations of the existing filtering channel optimization methods, avoid the optimization algorithm from falling into the optimal solution, and customize the filtering channel for the spectral characteristic peaks of different targets.

[0005] The present invention provides a method for optimizing the design of spectral filtering channels based on target characteristics. The filtering channel includes a substrate and layers of thin films stacked on the substrate in sequence, and includes the following steps:

[0006] Step 1. Collect the target spectral curve of the target; extract the number of characteristic peaks in the target spectral curve and number them in sequence;

[0007] Step 2. Select the material of the thin film according to the wavelength range of the target spectral curve;

[0008] Step 3. Randomly generate a candidate solution set containing candidate solutions for each layer of the thin film, expressed as:

[0009] ;

[0010] wherein, represents the refractive index of the thin film, represents the thickness of the thin film, represents the refractive index and thickness of the thin film of the th layer of the thin film and the th candidate solution;

[0011] Step 4. Randomly cross two by two the m candidate solutions of the same layer of the thin film to obtain m new candidate solutions. The m candidate solutions before crossing and the new m candidate solutions form a candidate solution set;

[0012] Step 5. Set the mutation rate to randomly change the material and thickness of all candidate solutions of one layer of the thin film in the candidate solution set obtained in Step 4 to obtain a mutated candidate solution set;

[0013] Step 6. Construct an objective function containing a penalty term, input the mutated candidate solution set into the objective function and screen out the individual with the minimum objective function value as the optimal filtering channel.

[0014] Compared with the prior art, the present application has the following advantages: The present invention generates m candidate solutions for each layer of the thin film of the filtering channel, combines genetic algorithms to perform crossover and mutation on the candidate solutions, expands the candidate solution set, eliminates the need to repeatedly debug the weights in the optimization algorithm, simplifies the computational complexity of the optimization algorithm, and then constructs an objective function containing a penalty term based on considering the diversity of target characteristic peaks to obtain the film system combination of the optimal filtering channel from the mutated candidate solution set, realizes free regulation of the central wavelength and bandwidth of the filtering channel, simultaneously prevents coupling overlap between multiple characteristic peaks of the same target, effectively overcomes the limitations of the existing optimization algorithms, and has stronger pertinence in the design results.

[0015] In a possible implementation manner, after the step 6, the following steps are further included:

[0016] Step 7. Introduce an error amount of 1% into the thickness of each layer of the thin film in the optimal filtering channel obtained in Step 6, and keep the material of each layer of the thin film unchanged;

[0017] Step 8. Calculate the objective function value of the optimal filtering channel after introducing the error amount, and set a threshold , and judge the objective function value to determine whether it is less than or equal to . If so, output the optimal filtering channel before introducing the error amount. The output optimal filtering channel includes the value and the transmittance curve of each layer of the thin film; if not, proceed to Step 9;

[0018] Step 9. Set the number of iterations , , set the maximum number of iterations , and judge whether it satisfies . If so, return to Step 3. If not, increase the number of candidate solutions in the candidate set, , and return to Step 3.

[0019] In a possible implementation manner, the formula for constructing the objective function including the penalty term in Step 6 is: :

[0020] ;

[0021] In the formula, represents the maximum wavelength of the filtering channel within the entire wavelength range, represents the minimum wavelength of the filtering channel within the entire wavelength range; represents the transmittance of the optimized filtering channel, represents the transmittance of the target spectral curve.

[0022] In a possible implementation manner, the formula for screening out the individual with the minimum objective function value as the optimal filtering channel based on the objective function in Step 6 is:

[0023] Minimize

[0024] ;

[0025] Subject to

[0026] ;

[0027] In the formula, Minimize represents minimization, Subject to represents constraints, is the number of the characteristic peak in the target spectral curve, represents the transmittance of the optimized filtering channel, represents the transmittance of the target spectral curve; represents the The maximum transmittance at each characteristic peak and its corresponding designed wavelength range; They are respectively the designed wavelength widths corresponding to 0.5 times, 0.8 times, and 0.05 times the maximum transmittance at the x th characteristic peak, indicating that when the transmittance at the

[0028] th characteristic peak is 0.5 times the maximum value, it is the corresponding target wavelength range. In a possible implementation manner, the transmittance of the optimized designed filter channel

[0029] is calculated as follows: k First, calculate the transfer matrix of the filter channel composed of M：

[0030] ;

[0031] In the formula, represents the phase thickness of the i th layer of film for the the j-th th candidate solution, 、 indicating the refractive index and the thickness refractive index of the th layer of film for the th candidate solution, represents the wavelength; indicates the optical admittance of the th layer of film for the th candidate solution, indicating the incident angle of the light beam passing through the th layer of film in the th candidate solution, represents the refractive index of the incident medium;

[0032] Then, the transmittance of the optimized designed filter channel is calculated by the formula:

[0033] . BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a schematic flow chart of the present invention;

[0035] Figure 2 is a structural diagram of the filter channel of the present invention;

[0036] Figure 3 is a comparison diagram of the design of the filter channel with double characteristic peaks and the target spectral curve in a specific embodiment of the present invention;

[0037] Figure 4Design parameters of the optimal filtering channel for the double characteristic peaks output by the specific embodiment of the present invention. Detailed implementation manners

[0038] First of all, those skilled in the art should understand that these implementation manners are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can adjust them as needed to adapt to specific application scenarios.

[0039] In the description of the embodiments of the present application, it should be noted that unless otherwise clearly specified and limited, the terms "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific situations.

[0040] In the embodiments of the present application, unless otherwise clearly specified and limited, the first feature being "on" or "under" the second feature can be that the first and second features are in direct contact, or the first and second features are indirectly in contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on" the second feature can be that the first feature is directly above or obliquely above the second feature, or merely indicates that the first feature has a higher horizontal height than the second feature. The first feature being "under", "below" and "beneath" the second feature can be that the first feature is directly below or obliquely below the second feature, or merely indicates that the first feature has a lower horizontal height than the second feature.

[0041] The present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] See Figure 1 , the embodiments of the present application disclose an optimization design method for a spectral filtering channel based on target features. As Figure 2 shown, the filtering channel includes a substrate and layers of thin films stacked on the substrate in sequence, and includes the following steps:

[0043] Step 1. Collect the target spectral curve of the target; extract the number of characteristic peaks in the target spectral curve and number them in sequence; the target of this specific embodiment is a leaf, and a spectrophotometer is used to obtain the target spectral curve of the leaf. As Figure 3 shown by the dotted line in, it is set that the transmittance in the wavelength range of 400 nm -445 nm and 481 nm -580 nm is 0, while in the wavelength range of 446 nm -480 nm and 581nm -590 nm The transmittance within the wavelength range is 1; and the wavelength range 446 nm -480 nm is extracted as characteristic peak 1, and the wavelength range 581 nm -590 nm is extracted as characteristic peak 2. Identifying characteristic peaks from the target spectral curve is prior art and will not be elaborated here;

[0044] Step 2. Select the material of the thin film according to the wavelength range of the target spectral curve; the wavelength range of the target spectral curve in this specific embodiment is 400 nm -700 nm , regarding the selection of the thin film material, follow the principle that the wavelength range of the thin film material falls within the wavelength range of the target spectral curve; the materials suitable for the wavelength range of 400 nm -700 nm in this specific embodiment are TiO2 and SiO2;

[0045] Step 3. Randomly generate a candidate solution set containing candidate solutions for each layer of the thin film, denoted as:

[0046] ;

[0047] wherein, represents the refractive index of the thin film, represents the thickness of the thin film, represents the refractive index and thickness of the thin film of the th layer of the thin film for the th candidate solution;

[0048] Step 4. Randomly cross the candidate solutions of the same layer of the thin film in pairs. For example, two candidate solutions , are crossed to generate new candidate solutions , ; in this specific embodiment, after randomly selecting two candidate solutions for crossing to generate new candidate solutions, randomly select two candidate solutions from the remaining candidate solutions to continue crossing. The 3000 candidate solutions before crossing and the 3000 new candidate solutions generated after crossing are added together to form a candidate solution set with 6000 candidate solutions;

[0049] Step 5. Set the mutation rate to 0.01 to randomly change the materials and thicknesses of all candidate solutions of one layer of the thin film in the candidate solution set obtained in Step 4 to obtain a mutated candidate solution set;

[0050] Step 6. Construct an objective function including a penalty term , and input the mutated candidate solution set into the objective function Among them, the individual with the smallest objective function value is selected as the optimal filtering channel; where the objective function The calculation formula is:

[0051] ;

[0052] In the formula, represents the maximum wavelength of the filtering channel within the entire wavelength range, represents the minimum wavelength of the filtering channel within the entire wavelength range; represents the transmittance of the optimized filtering channel, represents the transmittance of the target spectral curve;

[0053] Based on the objective function The calculation formula for selecting the individual with the smallest objective function value as the optimal filtering channel is:

[0054] Minimize

[0055] ;

[0056] Subject to

[0057] ;

[0058] In the formula, Minimize means minimization, and Subject to means constraint, is the number of the characteristic peak in the target spectral curve, represents the transmittance of the optimized filtering channel, represents the transmittance of the target spectral curve; represents the maximum transmittance and its corresponding designed wavelength range at the th characteristic peak; are respectively the designed wavelength widths corresponding to 0.5 times, 0.8 times, and 0.05 times of the maximum transmittance at the x th characteristic peak, represents the target wavelength range corresponding to 0.5 times of the maximum transmittance at the th characteristic peak.

[0059] The transmittance of the optimized filtering channel is calculated as follows:

[0060] First, calculate the transfer matrix k of the filtering channel composed of M：

[0061] ;

[0062] In the formula, Indicates the i phase thickness of the the j-th th layer of thin film and the 、 phase thickness of the th layer of thin film and the refractive index of the th candidate solution and the thickness refractive index of the thin film, Indicates the th layer of thin film and the optical admittance of the th candidate solution, Indicates the incident angle of the light beam passing through the th layer of thin film among the th candidate solutions,

[0063] Next, the transmittance of the optimized filter channel is calculated by the formula:

[0064] ;

[0065] Step 7. Introduce an error amount of 1% to the thickness of each layer of thin film in the optimal filter channel obtained in Step 6, and keep the material of each layer of thin film unchanged;

[0066] Step 8. Calculate the objective function value of the optimal filter channel after introducing the error amount, and set the threshold , and determine whether the objective function value is less than or equal to . If so, output the optimal filter channel before introducing the error amount. The output optimal filter channel includes the value of each layer of thin film and the transmittance curve; if not, go to Step 9;

[0067] Step 9. Set the number of iterations , , set the maximum number of iterations , and determine whether is satisfied. If so, return to Step 3. If not, expand the number of candidate solutions in the candidate set, , and return to Step 3.

[0068] The present invention generates m candidate solutions for each layer of thin film in the filter channel, combines the genetic algorithm to perform crossover and mutation on the candidate solutions, expands the candidate solution set, eliminates the need to repeatedly debug the weights in the optimization algorithm, simplifies the computational complexity of the optimization algorithm, and then constructs an objective function including a penalty term based on considering the diversity of target characteristic peaks The film system combination of the optimal filtering channel is obtained from the mutated candidate solution set, realizing the free regulation of the central wavelength and bandwidth of the filtering channel, while preventing the coupling overlap between multiple characteristic peaks of the same target, effectively overcoming the limitations of the existing optimization algorithms, and making the design result more targeted.

[0069] In the description of the embodiments of the present application, it should be noted that in the description of the present application, the terms indicating the direction or positional relationship such as "inner", "outer", etc. are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description, rather than indicating or implying that the device or component must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of the present application.

[0070] In the description of the present application, the description with reference to terms such as "one embodiment", "some embodiments", "in this embodiment", "specific examples", or "some examples" means that the specific features, mechanisms, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0071] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for optimizing the design of a spectral filtering channel based on target features, the filtering channel comprising a substrate and layers of thin films successively stacked on the substrate, characterized in that Including the following steps: Step 1. Collect the target spectral curve of the target; extract the number of characteristic peaks in the target spectral curve and number them in sequence; Step 2. Select the material of the thin film according to the wavelength range of the target spectral curve; Step 3. Randomly generate a candidate solution set containing candidate solutions for each layer of the thin film, expressed as: ； Wherein, represents the refractive index of the thin film, represents the thickness of the thin film, represents the refractive index and thickness of the thin film of the n-th candidate solution of the thin film layer; Step 4. For the same layer of films, randomly cross every two of the candidate solutions to obtain new candidate solutions. The candidate solutions before crossing and the new candidate solutions form a candidate solution set; Step 5. Set the mutation rate to randomly change the material and thickness of all candidate solutions of one of the thin films in the candidate solution set obtained in Step 4 to obtain a mutated candidate solution set; Step 6. Construct an objective function containing a penalty term , input the set of mutated candidate solutions into the objective function and select the individual with the minimum objective function value as the optimal filtering channel; The calculation formula of the objective function including the penalty term is: ； In the formula, represents the maximum wavelength of the filtering channel within the entire wavelength range, represents the minimum wavelength of the filtering channel within the entire wavelength range; represents the transmittance of the filtering channel after the optimized design, represents the transmittance of the target spectral curve; Based on the objective function The calculation formula for screening out the individual with the minimum objective function value as the optimal filtering channel is: ； In the formula, Minimize represents minimization, and Subject to represents constraint. is the number of the characteristic peak in the target spectral curve. represents the transmittance of the optimized filtering channel. represents the transmittance of the target spectral curve. represents the maximum transmittance at the -th characteristic peak and its corresponding designed wavelength range. 、 are respectively the designed wavelength widths corresponding to 0.5 times, 0.8 times and 0.05 times of the maximum transmittance at the -th characteristic peak. represents the target wavelength range corresponding to 0.5 times of the maximum transmittance at the -th characteristic peak. represents the maximum transmittance at the 1st characteristic peak. Step 7. Introduce an error amount of 1% into the thickness of each thin film in the optimal filtering channel obtained in Step 6 and keep the material of each thin film unchanged; Step 8. Calculate the objective function value of the optimal filtering channel after introducing the error amount and set a threshold , and determine the objective function value to see if it is less than or equal to . If so, output the optimal filtering channel before introducing the error amount. The output optimal filtering channel includes the , value and the transmittance curve of each layer of the thin film; if not, proceed to Step 9; Step 9. Set the number of iterations , , set the maximum number of iterations , and determine whether it meets , if yes, return to Step 3, if not, then increase the number of candidate solutions in the candidate set, , and return to Step 3.

2. The method for optimizing the design of a spectral filtering channel based on target features according to claim 1, wherein The transmittance of the optimized filtering channel The calculation includes: First calculate the transmission matrix of the filtering channel composed of thin films M : ； In the formula, represents the phase thickness of the th layer of the thin film for the th candidate solution, 、 represents the refractive index of the th layer of the thin film and the thickness refractive index of the thin film for the th candidate solution, represents the wavelength; represents the optical admittance of the th layer of the thin film for the th candidate solution, represents the incident angle of the light beam passing through the th layer of the thin film among the th candidate solution, and represents the refractive index of the incident medium; Next, the transmittance of the optimized filter channel The calculation formula is as follows: 。

Images