Single-pixel imaging-free target positioning method

Abstract

The invention provides a single-pixel imaging-free target positioning method, which comprises the following steps of S1, alternately loading a template pattern and a pure white pattern to the same position in a digital micromirror device by a computer, and respectively projecting the template pattern and the pure white pattern to an object pattern; s2, the light beam projected to the surface of the object is received by a single-pixel detector after being matched and reflected; and S3, constructing a target positioning model based on a genetic algorithm and template matching, dividing two single pixel values, and normalizing to serve as a new matching function for improving the template matching algorithm. S4, the genetic algorithm optimizes the fitness value obtained by each time of matching until the maximum number of iterations is reached or the solution with the highest fitness value is found; and S5, displaying the position of the template through the corresponding coordinate of the solution with the highest fitness value obtained in the matching process, and obtaining a to-be-positioned target. The algorithm and the system are both low in complexity, and meanwhile, compared with traditional template matching positioning, the method has higher operation speed.

Description

Technical Field

[0001] The present application relates to the field of imaging-free target positioning and single-pixel detection technology, and in particular to a single-pixel imaging-free target positioning method. Background Art

[0002] With the vigorous development of technologies such as autonomous driving and radar detection, the actual application needs of target positioning have gradually become diversified. People's requirements for imaging systems are not limited to improving resolution, but also include observing more comprehensive and complex target information in a larger spatial range, and accurately positioning the detected target. Especially in complex scenes, it is necessary to process targets in complex scenes in real time, and high-precision and low-cost target positioning technology is particularly important. At present, most of the existing target positioning methods are based on the "imaging first-positioning later" mode, first reconstructing the image of the target and then matching and positioning. However, when facing complex scenes, the imaging first and then positioning mode is prone to cause the target to be affected by environmental interference. When the reconstructed image resolution is insufficient or the target image is not clear, the positioning accuracy will be significantly reduced, and the pre-imaging process will also increase the system's computing time and memory. The existing "imaging first-positioning later" method is difficult to resist the influence of complex interference and exhibits poor robustness.

[0003] Computational optical imaging, which has been booming in recent years, combines technologies such as computers, optics, sensors, images, and signal processing, and has changed the "what you see is what you get" model of traditional cameras in terms of imaging mechanism. For example, unlike traditional array detectors, the light field acquisition method based on single-pixel detectors does not perform spatial resolution measurements, but only records the sum of the received light intensities, and uses the measured values ​​to reconstruct the image. Therefore, the image reconstruction algorithm is the key factor restricting the imaging quality. In order to improve the positioning accuracy, the existing "imaging first, positioning later" model has proposed three-dimensional reconstruction, compressed sensing, deep learning and other methods to improve the quality of image reconstruction, restore high-quality images and then locate the target. However, the "imaging first, positioning later" model combined with the above-mentioned image restoration and reconstruction methods usually increases the complexity of the system and increases the running time.

[0004] Application Contents

[0005] The present application is made in view of the above-mentioned problems, and its purpose is to provide a single-pixel imaging-free target positioning method, combining template matching and genetic algorithm principles, and proposing a new normalized matching function. This method can perform target positioning when the target to be positioned is roughly known but the entire image scene is unknown. Image reconstruction is not required, and the algorithm and system complexity are relatively low. At the same time, it has a faster running speed than traditional template matching positioning.

[0006] Specifically, the first aspect of the present application provides a single-pixel imaging-free target positioning method, comprising the following steps:

[0007] S1: The computer alternately loads the template pattern and the pure white pattern into the same position in the digital micromirror device, and the light beam emitted by the light source irradiates the digital micromirror device and is projected onto the object pattern respectively;

[0008] Digital micromirror device (DMD) is a micro-electromechanical system with electronic input and optical output. It is composed of many small aluminum reflective mirrors, each of which is called a pixel. Each mirror can deflect ±12° around the diagonal of each small mirror in the positive direction. It has high resolution, high contrast and fast response time.

[0009] S2: The light beam projected onto the surface of the object is matched and reflected before being received by a single-pixel detector;

[0010] A single pixel detector is used to receive light signals and calculate single pixel values.

[0011] S3: Build a target positioning model based on genetic algorithm and template matching, calculate the single pixel value of the area covered by the template and the pure white pattern, divide the two single pixel values ​​and normalize them as a new matching function of the improved template matching algorithm, and also as the fitness function of the genetic algorithm;

[0012] The formula of the new matching function is as follows:

[0013]

[0014] Where R(x, y): the similarity between the template image and the target image at point (x, y);

[0015] T(x′, y′): pixel value of the template;

[0016] The pixel mean of the template; The pixel mean of the target image;

[0017] I(x+′, y+y′): pixel value of the template in the corresponding area of ​​the target area;

[0018] ∑ x，y I(x, y): the sum of the pixel values ​​at the corresponding position of the target image;

[0019] Covariance of image and template;

[0020] The fitness function of the genetic algorithm is an indicator used to judge the quality of individuals in the group, which directly affects the performance of the genetic algorithm.

[0021] S4: The genetic algorithm optimizes the fitness value obtained from each match: repeating the selection, crossover and mutation steps, retaining the function value with a high fitness value until the maximum number of iterations is reached or the solution with the highest fitness value is found;

[0022] S5: The corresponding coordinates of the solution with the highest fitness value obtained during the matching process are used to display the position of the template. The target to be located can be obtained by selecting a range of the same size as the template in the target image.

[0023] Furthermore, the computer alternately loads the template pattern and the pure white pattern into the same position in the digital micromirror device, specifically including:

[0024] The computer sets the template to be projected and the pure white pattern to be 128*128 pixels in size. The pixel value of the pure white pattern is 1. The result of convolution of the pure white pattern with the same size area in the target pattern is the same area in the target pattern itself. Therefore, in the experiment, it is equivalent to the pixel value of the area. Then, the template pattern with a size of 128*128 pixels is loaded into the digital micromirror device.

[0025] Furthermore, the light beam emitted by the light source irradiates the digital micromirror device and is projected onto the object pattern respectively, specifically including: the light source emits structured light, irradiates the surface of the digital micromirror device, and after reflection and convergence with the lens, completely irradiates the surface of the target object.

[0026] Furthermore, the light beam projected onto the surface of the object is received by the single-pixel detector after matching and reflection, which specifically includes: placing a lens between the digital micromirror device and the single-pixel detector, and utilizing the convergence function of the lens to adjust the distance between the digital micromirror device, the lens and the single-pixel detector, so that the light beam reflected from the target image surface can be completely received by the single-pixel detector after convergence by the lens.

[0027] Furthermore, the lens functions to ensure that the light field and the target pattern are of the same size and that the detector can completely receive the light signal.

[0028] Furthermore, the target positioning model based on the genetic algorithm and template matching is constructed, the single pixel value of the area covered by the template and the pure white pattern is calculated, and the two single pixel values ​​are divided and normalized as a new matching function of the improved template matching algorithm, and also as a fitness function of the genetic algorithm, including the following steps:

[0029] S11: First, determine the template size to be 128*128 pixels and the target image size to be 400*400 pixels. In the experiment, to facilitate receiving light intensity, the template is embedded in a 400*400 pixel matrix, and the corresponding pixels of the template are set to 1, and the rest are 0; slide the template and calculate the similarity between the template and the corresponding area of ​​the target image;

[0030] S12: The genetic algorithm defines a fitness function to evaluate the quality of each individual. It measures the matching degree between the template and the target to be located according to the matching value obtained from each template matching, performs global optimization, reduces the number of line-by-line scanning and matching, improves the running time, and thus finds the position with the highest fitness value more quickly.

[0031] S13: First, the template pattern is matched with the target pattern, that is, the corresponding areas are convolved and summed, and then the sum of the corresponding areas is calculated, the two single pixel values ​​are divided, and normalized as the improved fitness function of the genetic algorithm.

[0032] Furthermore, the template pattern is matched with the target pattern, and a process of calculating a single pixel value of an overlapping area between the template pattern and the target pattern surface includes the following steps:

[0033] S21: The template pattern size is 128*128 pixels, which is embedded in a 400*400 pixel matrix equal to the target image size. The corresponding pixels of the template are set to 1 and the rest are set to 0, which serves as the convolution kernel.

[0034] S22: Convolution calculation is performed on the corresponding area of ​​the template and the target image, that is, the light intensity value obtained by multiplying and adding the corresponding pixel values ​​is received by the single pixel detector after reflection and is a single pixel value.

[0035] Furthermore, the optimization process of the genetic algorithm comprises the following steps:

[0036] S31: taking each matching position as an individual, defining the population size and the encoding length of the individual, using binary encoding, generating a matrix with the number of rows equal to the population size and the number of columns equal to the encoding length, wherein the elements are randomly generated 0 and 1, representing the binary encoding of the individual;

[0037] S32: Perform selection operation, select individuals with larger fitness function values, and replicate the next generation;

[0038] S33: Perform a crossover operation to exchange some genes of two individual code strings with a certain crossover probability Pc to generate two new individuals;

[0039] S34: Perform mutation operation, with the mutation probability Pm replacing some gene values ​​in the individual code string with other gene values, thereby forming a new individual and maintaining diversity;

[0040] S35: Repeat S32-S34 and stop when the number of iterations or the fitness value reaches the maximum.

[0041] Furthermore, the matching result is normalized to a range between [-1, 1], where 1 indicates a complete match and -1 indicates a complete mismatch.

[0042] Furthermore, the corresponding coordinates of the solution with the highest fitness value obtained in the matching process, showing the position of the template, include the following steps:

[0043] S41: During the optimization process, each time the template is matched, the position of the template in the 400*400 pixel matrix is ​​recorded, and the optimization and matching are stopped until the target position is found;

[0044] S42: When the target position is found, the optimal fitness value is obtained, and the final position of the template in the 400*400 pixel matrix is ​​obtained;

[0045] S43: The optimal fitness value corresponds to the pixel coordinates at the upper left corner. Taking the coordinates as the starting point, a rectangular box of 128*128 pixels is drawn. The selected area is the target to be located.

[0046] In the second aspect, the present application also provides a computing device, which has the function of implementing the method described in the first aspect above. The beneficial effects can be found in the description of the first aspect, which will not be repeated here. The function can be implemented by hardware, or by executing the corresponding software implementation through hardware. The hardware or software includes one or more modules corresponding to the above functions. In one possible design, the structure of the device includes an acquisition module, a training module, and optionally, a construction module. These modules can implement the function of the training node in the method example of the first aspect above. Please refer to the detailed description in the method example for details, which will not be repeated here.

[0047] In the third aspect, the present application also provides a computing device, which is used to implement the functions of the method described in the first aspect above. The beneficial effects can be found in the description of the first aspect and will not be repeated here. The structure of the computing device includes a processor and a memory, and the memory is used to store instructions and / or data. The memory is coupled to the processor, and when the processor executes the program instructions stored in the memory, the function of the training node in the example of the first aspect above can be implemented. The structure of the computing device also includes a communication interface for communicating with other devices.

[0048] In a fourth aspect, the present application also provides a computer-readable storage medium, in which instructions are stored, and when the computer-readable storage medium is run on a computer, the computer executes the method in the above-mentioned first aspect and various possible designs of the first aspect.

[0049] In a fifth aspect, the present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the method in the above-mentioned first aspect and various possible designs of the first aspect.

[0050] In a sixth aspect, the present application also provides a computing chip, which is connected to a memory, and the chip is used to read and execute a software program stored in the memory, and to execute the methods in the above-mentioned first aspect and various possible implementation methods of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present drawings or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present drawings. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0052] Figure 1 A flowchart of a single-pixel imaging-free target positioning method;

[0053] Figure 2 It is a schematic diagram of the overall composition of the present invention;

[0054] Figure 3 This is a flow chart of the genetic algorithm for improving the fitness function of the present invention.

[0055] The purpose, features and advantages of this figure will be further described in conjunction with the embodiments with reference to the accompanying drawings. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0057] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0058] Unless otherwise specified, all embodiments and optional embodiments of the present application can be combined with each other to form a new technical solution.

[0059] Unless otherwise specified, all technical features and optional technical features of this application can be combined with each other to form a new technical solution.

[0060] If there is no special explanation, all steps of the present application can be performed sequentially or randomly, preferably sequentially. For example, the method includes steps (a) and (b), which means that the method may include steps (a) and (b) performed sequentially, or may include steps (b) and (a) performed sequentially. For example, the method may further include step (c), which means that step (c) may be added to the method in any order. For example, the method may include steps (a), (b) and (c), or may include steps (a), (c) and (b), or may include steps (c), (a) and (b), etc.

[0061] If there is no special explanation, the "include" and "comprising" mentioned in this application are open-ended or closed-ended. For example, the "include" and "comprising" may mean that other components not listed may also be included or only the listed components may be included or only the listed components may be included.

[0062] If not specifically stated, in this application, the term "or" is inclusive. For example, the phrase "A or B" means "A, B, or both A and B". More specifically, any of the following conditions satisfies the condition "A or B": A is true (or exists) and B is false (or does not exist); A is false (or does not exist) and B is true (or exists); or both A and B are true (or exist).

[0063] In order to better understand the solutions of the embodiments of the present application, some relevant terms and concepts that may be involved in the embodiments of the present application are first introduced below.

[0064] (1) Template matching refers to finding the part of the test image that is similar to the template image by comparing the template image with the test image. This is achieved by calculating the similarity between the template image and the target in the test image. It can quickly locate the predefined target in the test image.

[0065] (2) Genetic Algorithm (GA), which is designed and proposed based on the evolutionary laws of organisms in nature. It is a computational model of biological evolution that simulates the natural selection and genetic mechanisms of Darwin's theory of biological evolution. It is a method of searching for the optimal solution by simulating the natural evolution process. The algorithm uses mathematical methods and computer simulation operations to convert the problem-solving process into a process similar to the crossover and mutation of chromosome genes in biological evolution. When solving more complex combinatorial optimization problems, it can usually obtain better optimization results faster than some conventional optimization algorithms. Genetic algorithms have been widely used in combinatorial optimization, machine learning, signal processing, adaptive control, artificial life and other fields.

[0066] (3) Fitness function. The fitness function of a genetic algorithm is also called an evaluation function. It is an indicator used to judge the quality of individuals in a group. It is evaluated based on the objective function of the problem being solved. Genetic algorithms generally do not require other external information during the search and evolution process. They only use evaluation functions to evaluate the quality of individuals or solutions and serve as the basis for subsequent genetic operations. Since the fitness function in a genetic algorithm compares and sorts and calculates the selection probability on this basis, the value of the fitness function must be positive. It can be seen that in many cases, it is necessary to map the objective function into a fitness function in the form of maximum value and with a non-negative function value. The design of the fitness function should be based on the requirements of the problem being solved. The design of the fitness function directly affects the performance of the genetic algorithm.

[0067] like Figure 1 As shown, a single-pixel imaging-free target positioning method comprises the following steps:

[0068] S1: The computer alternately loads the template pattern and the pure white pattern into the same position in the digital micromirror device, and the light beam emitted by the light source irradiates the digital micromirror device and is projected onto the object pattern respectively;

[0069] A digital micromirror device is a micro-electromechanical system with electronic input and optical output. It is composed of many small aluminum reflective mirrors, each of which is called a pixel. Each mirror can deflect ±12° around the diagonal of each small mirror in the positive direction. It has high resolution, high contrast and fast response time.

[0070] S2: The light beam projected onto the surface of the object is matched and reflected before being received by a single-pixel detector;

[0071] A single pixel detector is used to receive light signals and calculate single pixel values.

[0072] S3: Build a target positioning model based on genetic algorithm and template matching, calculate the single pixel value of the area covered by the template and the pure white pattern, divide the two single pixel values ​​and normalize them as a new matching function of the improved template matching algorithm, and also as the fitness function of the genetic algorithm;

[0073] The formula of the new matching function is as follows:

[0074]

[0075] The fitness function of the genetic algorithm is an indicator used to judge the quality of individuals in the group, which directly affects the performance of the genetic algorithm.

[0076] S4: The genetic algorithm optimizes the fitness value obtained from each match: repeating the selection, crossover and mutation steps, retaining the function value with a high fitness value until the maximum number of iterations is reached or the solution with the highest fitness value is found;

[0077] The purpose of the selection operation is to directly pass on the optimized individuals to the next generation or to generate new individuals through pairing and crossover and then pass on to the next generation. The selection operation is based on the fitness evaluation of individuals in the group.

[0078] Crossover operation refers to the operation of replacing and recombining parts of the structures of two parent individuals to generate new individuals. Through crossover, the search ability of genetic algorithms is greatly improved.

[0079] The basic content of mutation operation is to change the gene values ​​on certain loci of individual strings in the population. According to different individual coding representation methods. The purpose of mutation operation is to enable the genetic algorithm to have local random search capabilities. When the genetic algorithm has approached the neighborhood of the optimal solution through the crossover operator, the local random search ability of the mutation operator can accelerate the convergence to the optimal solution. Obviously, the mutation probability in this case should be smaller, otherwise the building blocks close to the optimal solution will be destroyed due to mutation. Second, the genetic algorithm can maintain population diversity to prevent premature convergence. At this time, the convergence probability should be larger.

[0080] S5: The corresponding coordinates of the solution with the highest fitness value obtained during the matching process are used to display the position of the template. The target to be located can be obtained by selecting a range of the same size as the template in the target image.

[0081] Furthermore, the computer alternately loads the template pattern and the pure white pattern into the same position in the digital micromirror device, specifically including:

[0082] The computer sets the template to be projected and the pure white pattern to be 128*128 pixels in size. The pixel value of the pure white pattern is 1. The result of convolution of the pure white pattern with the same size area in the target pattern is the same area in the target pattern itself. Therefore, in the experiment, it is equivalent to the pixel value of the area. Then, the template pattern with a size of 128*128 pixels is loaded into the digital micromirror device.

[0083] Furthermore, the light beam emitted by the light source irradiates the digital micromirror device and is projected onto the object pattern respectively, specifically including: the light source emits structured light, irradiates the surface of the digital micromirror device, and after reflection and convergence with the lens, completely irradiates the surface of the target object.

[0084] In this embodiment, the light source used is a projector, which is used to emit structured light;

[0085] Furthermore, the light beam projected onto the surface of the object is received by the single-pixel detector after matching and reflection, which specifically includes: placing a lens between the digital micromirror device and the single-pixel detector, and utilizing the convergence function of the lens to adjust the distance between the digital micromirror device, the lens and the single-pixel detector, so that the light beam reflected from the target image surface can be completely received by the single-pixel detector after convergence by the lens.

[0086] Furthermore, the lens functions to ensure that the light field and the target pattern are of the same size and that the detector can completely receive the light signal.

[0087] Furthermore, the target positioning model based on the genetic algorithm and template matching is constructed, the single pixel value of the area covered by the template and the pure white pattern is calculated, and the two single pixel values ​​are divided and normalized as a new matching function of the improved template matching algorithm, and also as a fitness function of the genetic algorithm, including the following steps:

[0088] S11: First, determine the template size to be 128*128 pixels and the target image size to be 400*400 pixels. In the experiment, to facilitate receiving light intensity, the template is embedded in a 400*400 pixel matrix, and the corresponding pixels of the template are set to 1, and the rest are 0; slide the template and calculate the similarity between the template and the corresponding area of ​​the target image;

[0089] S12: The genetic algorithm defines a fitness function to evaluate the quality of each individual. It measures the matching degree between the template and the target to be located according to the matching value obtained from each template matching, performs global optimization, reduces the number of line-by-line scanning and matching, improves the running time, and thus finds the position with the highest fitness value more quickly.

[0090] S13: First, the template pattern is matched with the target pattern, that is, the corresponding areas are convolved and summed, and then the sum of the corresponding areas is calculated, the two single pixel values ​​are divided, and normalized as the improved fitness function of the genetic algorithm.

[0091] Furthermore, the template pattern is matched with the target pattern, and a process of calculating a single pixel value of an overlapping area between the template pattern and the target pattern surface includes the following steps:

[0092] S21: The template pattern size is 128*128 pixels, which is embedded in a 400*400 pixel matrix equal to the target image size. The corresponding pixels of the template are set to 1 and the rest are set to 0, which serves as the convolution kernel.

[0093] S22: Convolution calculation is performed on the corresponding area of ​​the template and the target image, that is, the light intensity value obtained by multiplying and adding the corresponding pixel values ​​is received by the single pixel detector after reflection and is a single pixel value.

[0094] Furthermore, the optimization process of the genetic algorithm comprises the following steps:

[0095] S31: taking each matching position as an individual, defining the population size and the encoding length of the individual, using binary encoding, generating a matrix with the number of rows equal to the population size and the number of columns equal to the encoding length, wherein the elements are randomly generated 0 and 1, representing the binary encoding of the individual;

[0096] S32: Perform selection operation, select individuals with larger fitness function values, and replicate the next generation;

[0097] S33: Perform a crossover operation to exchange some genes of two individual code strings with a certain crossover probability Pc to generate two new individuals;

[0098] S34: Perform mutation operation, with the mutation probability Pm replacing some gene values ​​in the individual code string with other gene values, thereby forming a new individual and maintaining diversity;

[0099] S35: Repeat S32-S34 and stop when the number of iterations or the fitness value reaches the maximum.

[0100] Furthermore, the matching result is normalized to a range between [-1, 1], where 1 indicates a complete match and -1 indicates a complete mismatch.

[0101] Furthermore, the corresponding coordinates of the solution with the highest fitness value obtained in the matching process, showing the position of the template, include the following steps:

[0102] S41: During the optimization process, each time the template is matched, the position of the template in the 400*400 pixel matrix is ​​recorded, and the optimization and matching are stopped until the target position is found;

[0103] S42: When the target position is found, the optimal fitness value is obtained, and the final position of the template in the 400*400 pixel matrix is ​​obtained;

[0104] S43: The optimal fitness value corresponds to the pixel coordinates at the upper left corner. Taking the coordinates as the starting point, a rectangular box of 128*128 pixels is drawn. The selected area is the target to be located.

[0105] It should be noted that the present application is not limited to the above-mentioned embodiments. The above-mentioned embodiments are only examples, and the embodiments having the same structure as the technical idea and exerting the same effect within the scope of the technical solution of the present application are all included in the technical scope of the present application. In addition, without departing from the scope of the main purpose of the present application, various modifications that can be thought of by those skilled in the art to the embodiments and other methods of combining some of the constituent elements in the embodiments are also included in the scope of the present application.

Claims

1. A single-pixel imaging-free target positioning method, It is characterized in that The following steps are involved: S1: The computer alternately loads the template pattern and the pure white pattern into the same position in the digital micromirror device, and the light beam emitted by the light source irradiates the digital micromirror device and is projected onto the object pattern respectively; S2: The light beam projected onto the surface of the object is matched and reflected before being received by a single-pixel detector; S3: Build a target positioning model based on genetic algorithm and template matching, calculate the single pixel value of the area covered by the template and the pure white pattern, divide the two single pixel values ​​and normalize them as a new matching function of the improved template matching algorithm, and also as the fitness function of the genetic algorithm; The formula of the new matching function is as follows: Where R(x, y): the similarity between the template image and the target image at point (x, y); T(x′, y′): pixel value of the template; The pixel mean of the template; The pixel mean of the target image; I(x+x′, y+y′): pixel value of the template in the corresponding area of ​​the target area; ∑ x，y I(x, y): the sum of the pixel values ​​at the corresponding position of the target image; Covariance of image and template; S4: The genetic algorithm optimizes the fitness value obtained from each match: repeating the selection, crossover and mutation steps, retaining the function value with a high fitness value until the maximum number of iterations is reached or the solution with the highest fitness value is found; S5: The corresponding coordinates of the solution with the highest fitness value obtained during the matching process are used to display the position of the template. The target to be located can be obtained by selecting a range of the same size as the template in the target image.

2. A single-pixel imaging-free target positioning method according to claim 1, It is characterized in that The computer alternately loads the template pattern and the pure white pattern into the same position of the digital micromirror device, specifically including: The computer sets the template to be projected and the pure white pattern to be 128*128 pixels in size. The pixel value of the pure white pattern is 1. The result of convolution of the pure white pattern with the same size area in the target pattern is the same area in the target pattern itself. Therefore, in the experiment, it is equivalent to the pixel value of the area. Then, the template pattern with a size of 128*128 pixels is loaded into the digital micromirror device.

3. A single-pixel imaging-free target positioning method according to claim 1, It is characterized in that The light beam emitted by the light source irradiates the digital micromirror device and is projected onto the object pattern respectively, which specifically includes: the light source emits structured light, irradiates the surface of the digital micromirror device, and completely irradiates the surface of the target object after being reflected and converged by the lens.

4. A single-pixel imaging-free target positioning method according to claim 1, It is characterized in that The light beam projected onto the surface of the object is matched and reflected and then received by the single-pixel detector, which specifically includes: placing a lens between the digital micromirror device and the single-pixel detector, and utilizing the convergence function of the lens to adjust the distance between the digital micromirror device, the lens and the single-pixel detector, so that the light beam reflected from the target image surface can be completely received by the single-pixel detector after being converged by the lens.

5. A single-pixel imaging-free target positioning method according to claim 4, It is characterized in that The function of the lens is to ensure that the light field and the target pattern are of the same size and that the detector can fully receive the light signal.

6. A single-pixel imaging-free target positioning method according to claim 1, It is characterized in that The target positioning model based on genetic algorithm and template matching is constructed, the single pixel value of the area covered by the template and the pure white pattern is calculated, and the two single pixel values ​​are divided and normalized as a new matching function of the improved template matching algorithm, and also as a fitness function of the genetic algorithm, including the following steps: S11: First, determine the template size to be 128*128 pixels and the target image size to be 400*400 pixels. In the experiment, to facilitate receiving light intensity, the template is embedded in a 400*400 pixel matrix, and the corresponding pixels of the template are set to 1, and the rest are 0; slide the template and calculate the similarity between the template and the corresponding area of ​​the target image; S12: The genetic algorithm defines a fitness function to evaluate the quality of each individual. It measures the matching degree between the template and the target to be located according to the matching value obtained from each template matching, performs global optimization, reduces the number of line-by-line scanning and matching, improves the running time, and thus finds the position with the highest fitness value more quickly. S13: First, the template pattern is matched with the target pattern, that is, the corresponding areas are convolved and summed, and then the sum of the corresponding areas is calculated, the two single pixel values ​​are divided, and normalized as the improved fitness function of the genetic algorithm.

7. A single-pixel imaging-free target positioning method according to claim 6, It is characterized in that The template pattern is matched with the target pattern, and the calculation process of the single pixel value of the overlapping area of ​​the template pattern and the target pattern surface includes the following steps: S21: The template pattern size is 128*128 pixels, which is embedded in a 400*400 pixel matrix equal to the target image size. The corresponding pixels of the template are set to 1 and the rest are set to 0, which serves as the convolution kernel. S22: Convolution calculation is performed on the corresponding area of ​​the template and the target image, that is, the light intensity value obtained by multiplying and adding the corresponding pixel values ​​is received by the single pixel detector after reflection and is a single pixel value.

8. A single-pixel imaging-free target positioning method according to claim 6, It is characterized in that The optimization process of the genetic algorithm comprises the following steps: S31: taking each matching position as an individual, defining the population size and the encoding length of the individual, using binary encoding, generating a matrix with the number of rows equal to the population size and the number of columns equal to the encoding length, wherein the elements are randomly generated 0 and 1, representing the binary encoding of the individual; S32: Perform selection operation, select individuals with larger fitness function values, and replicate the next generation; S33: Perform a crossover operation to exchange some genes of two individual code strings with a certain crossover probability Pc to generate two new individuals; S34: Perform mutation operation, with the mutation probability Pm replacing some gene values ​​in the individual code string with other gene values, thereby forming a new individual and maintaining diversity; S35: Repeat S32-S34 and stop when the number of iterations or the fitness value reaches the maximum.

9. A single-pixel imaging-free target positioning method according to claim 6, It is characterized in that The normalized results are matched in the range of [-1, 1], where 1 indicates a complete match and -1 indicates a complete mismatch.

10. The single-pixel imaging-free target positioning method according to claim 1, It is characterized in that The corresponding coordinates of the solution with the highest fitness value obtained in the matching process, showing the position of the template, include the following steps: S41: During the optimization process, each time the template is matched, the position of the template in the 400*400 pixel matrix is ​​recorded, and the optimization and matching are stopped until the target position is found; S42: When the target position is found, the optimal fitness value is obtained, and the final position of the template in the 400*400 pixel matrix is ​​obtained; S43: The optimal fitness value corresponds to the pixel coordinates at the upper left corner. Taking the coordinates as the starting point, a rectangular box of 128*128 pixels is drawn. The selected area is the target to be located.

Images