Self-adaptive sampling dynamic target tracking method and device based on single-pixel imaging
Through adaptive sampling and Fourier linear phase shift characteristic calculation, a cosine-cosine-coded illumination pattern is generated, which solves the efficiency and quality limitations of traditional single-pixel imaging in dynamic target tracking, and achieves efficient and economical dynamic target tracking.
Patent Information
- Application Number
- CN202510666381.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-22
- Publication Date
- 2025-08-15
AI Technical Summary
The traditional fixed sampling mode has limitations in efficiency and quality in dynamic target tracking, especially in high-speed moving targets, complex backgrounds and lighting changes.
Adaptive sampling method based on single-pixel imaging is adopted to generate four sets of cosine-coded illumination patterns with different phase characteristics, projected to the target scene through a spatial light modulator, and received reflected light signals in combination with the photodetector, and calculated the target displacement using the Fourier linear phase shift characteristics. The adaptive sampling strategy is used to dynamically adjust the sampling frequency, and the background frame difference method is used to eliminate static background interference.
It significantly improves the efficiency and image quality of single-pixel imaging, reduces hardware complexity and cost, and achieves fast and accurate dynamic target tracking, which is suitable for real-time monitoring in complex environments.
Smart Images
Figure CN120491098A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of single-pixel computational imaging technology, and more specifically, to a method and device for adaptive sampling dynamic target tracking based on single-pixel imaging. Background Art
[0002] Object tracking is a key technology in fields such as computer vision, artificial intelligence, and robotics. It is especially crucial in applications that require real-time response and decision-making. In dynamic scenes, the target's motion, posture changes, and environmental complexity all pose huge challenges to target tracking.
[0003] Traditional target tracking methods often rely on multi-pixel sensors and complex computational models. While they can achieve good results in certain scenarios, they still have many shortcomings when faced with high-speed moving targets, complex backgrounds, and changing lighting conditions. Single-pixel imaging technology provides a new approach to dynamic target tracking. As an emerging computational imaging technology, it does not rely on traditional multi-pixel sensors and only requires a single light sensor for acquisition. It uses a single-pixel detector combined with structured light illumination and advanced image reconstruction algorithms to achieve target imaging and tracking. Its design has lower design costs and a simpler structure. Unlike traditional imaging technologies, single-pixel imaging technology uses a serial sampling method, which has stronger anti-interference capabilities in environments with poor lighting conditions. Due to its advantages such as wide operating spectrum, high sensitivity, and non-line-of-sight imaging, single-pixel imaging technology has unique advantages in infrared, ultraviolet, terahertz bands, as well as in imaging through scattering media and non-line-of-sight, making it suitable for target tracking tasks in a variety of complex environments.
[0004] Single-pixel computational imaging for tracking dynamic targets requires a computer to obtain an illumination pattern through a modulation algorithm. The illumination pattern is then projected onto the target scene through a spatial light modulator. The reflected light is used by a photodetector to obtain the light intensity signal value. Finally, the photoelectric signal is converted and the computer performs data acquisition and analysis to obtain the required information. The target's motion is analyzed without an image. The illumination patterns include randomly generated illumination patterns and digitally encoded illumination patterns. The reconstruction process of random illumination patterns is relatively cumbersome, computationally intensive, and has large reconstruction errors. Summary of the Invention
[0005] In order to solve the problems of efficiency and quality limitations of traditional fixed sampling modes, the present application provides a method and device for dynamic target tracking with adaptive sampling based on single-pixel imaging.
[0006] The embodiment of the present application is implemented as follows:
[0007] In a first aspect, the present application provides a method for adaptive sampling dynamic target tracking based on single-pixel imaging, comprising:
[0008] Generate four sets of sine-cosine coded illumination patterns with different phase characteristics;
[0009] cyclically projecting the illumination pattern onto a target scene through a spatial light modulator to obtain a modulated signal;
[0010] The photoelectric detector receives the modulated reflected light signal of the target scene and converts the light signal into an electrical signal, which is received by the data acquisition card;
[0011] Adopting adaptive sampling strategy to dynamically adjust the sampling frequency, and controlling the maximum and minimum sampling times based on the voltage difference threshold;
[0012] The target displacement is calculated using the Fourier linear phase shift characteristic, and the background frame difference method is combined to eliminate static background interference and achieve real-time tracking of dynamic targets.
[0013] In a possible implementation, the formula for generating the lighting pattern includes:
[0014] P1=m+ncos(2πf x x+2πf y y+0);
[0015]
[0016] P3=m+ncos(2πf x x+2πf y y+π);
[0017]
[0018] Among them, m is the average light intensity, n is the contrast, x and y are the coordinates of the plane where the target object is located, and f x and f y are the spatial frequencies in the x and y directions respectively.
[0019] In a possible implementation, the lighting pattern includes two sets of spatial frequency combinations including:
[0020] Group 1: f x =2 / 128, f y =0;
[0021] Group 2: f x =0,f y =2 / 128;
[0022] Each group contains four phases with different phases. A total of eight lighting patterns are projected in a loop.
[0023] 5. In one possible implementation, the rules of the adaptive sampling strategy are:
[0024]
[0025] Where |Vi-Vi-1| represents the voltage difference between the i-th sample and the i-1-th sample, T is the voltage change threshold, and Nimax and Nimin are the dynamically adjusted maximum and minimum sampling times, respectively. These values increase or decrease based on voltage fluctuations, controlling the sampling range and ensuring increased sampling during rapid changes and decreased sampling during slow changes, thereby balancing resource usage and accuracy. Sfast and Sslow are the sample count adjustment steps for fast and slow sampling, respectively.
[0026] In one possible implementation, the adaptive sampling strategy is used when collecting data, increasing the sampling rate when the target object moves faster and reducing the sampling rate when the target object moves slower, so as to achieve a balance between data utilization and system efficiency, thereby reducing unnecessary data redundancy.
[0027] In a possible implementation, the target displacement is calculated as follows:
[0028]
[0029] Among them, angle{} represents the operation of parameters, I represents the image of the target object, and I b Represents the coefficients obtained from the background frame, which is the frame obtained before the target object enters the scene or starts moving. x0 and y0 are the displacements calculated in the x and y directions, respectively.
[0030] In one possible implementation, the formula for obtaining the target scene is:
[0031] F = (D1-D3) + j·(D2-D4);
[0032] I(x-x0,y-y0)=F -1 {I(f x ,f y )exp[-j2π(f x x0+f x y0)]};
[0033] Among them, D1, D2, D3, and D4 are the electrical signals corresponding to P1, P2, P3, and P4 in the above formula, respectively, and are also the electrical signals converted by the photoelectric unit. F represents the result of Fourier transform, which converts the image from the spatial domain to the frequency domain through a four-step phase shift method. j represents the sign of the imaginary part.
[0034] In a possible implementation, the spatial light modulator is a DLP projector or an LCD display screen, which is used to project the sine-cosine coded illumination pattern, and the photodetector is a photodiode, which receives the reflected light signal and outputs an electrical signal through a photoelectric conversion unit.
[0035] In a possible implementation, the method further includes:
[0036] The reconstructed image is projected back into the scene via a spatial light modulator, enabling intuitive presentation of the target object's position information.
[0037] The changes in the pixels in the final reconstructed pattern directly reflect the location information of the target object;
[0038] By comparing the calculation results of different pixel points, the specific position of the target object in the scene can be determined quickly and accurately.
[0039] In a second aspect, the present application provides a dynamic target tracking device with adaptive sampling based on single-pixel imaging, comprising:
[0040] a pattern generation module, configured to generate four sets of sine-cosine coded illumination patterns with different phase characteristics;
[0041] a signal modulation module, configured to cyclically project the illumination pattern onto a target scene via a spatial light modulator to obtain a modulated signal;
[0042] A signal acquisition module is used to receive the modulated reflected light signal of the target scene based on a photoelectric detector and convert the light signal into an electrical signal to be received by a data acquisition card;
[0043] Sampling frequency adjustment module, used to dynamically adjust the sampling frequency using an adaptive sampling strategy and control the maximum and minimum sampling times based on the voltage difference threshold;
[0044] The displacement calculation module is used to calculate the target displacement using the Fourier linear phase shift characteristics, and combine the background frame difference method to eliminate static background interference and realize real-time tracking of dynamic targets.
[0045] The technical solution provided by this application can achieve at least the following beneficial effects:
[0046] This application uses four sets of sine or cosine coding patterns with different phases to efficiently achieve image reconstruction and target object tracking, significantly reducing the number of required illumination patterns and greatly improving the efficiency of single-pixel imaging. At the same time, the application of the four-step phase shifting method effectively improves the image quality, making the reconstructed image clearer and more detailed, providing a solid foundation for accurate tracking.
[0047] In terms of hardware application, the spatial light modulator used in this application only needs to project the illumination pattern once into the scene for modulation to obtain the light intensity signal required for the entire experimental process. There is no need to perform multiple measurements, which greatly improves the experimental efficiency and simplifies the operation process.
[0048] This application uses single-pixel computational imaging technology to track target objects, which can be completed by relying solely on a single-pixel imaging system without the need for additional complex monitoring tools. This significantly reduces monitoring costs and shortens processing time, making dynamic target tracking more efficient and economical.
[0049] This application uses a differential method to calculate displacement, using the Fourier linear phase shift formula to calculate the displacement difference between the current frame and the previous frame, thereby accurately obtaining the displacement of the target object in the scene. This method can quickly obtain results without extensive analysis and processing, and can intuitively display the target object's motion by reconstructing the pixel changes in the image, combining efficiency and practicality.
[0050] This application uses an adaptive sampling strategy to analyze target motion characteristics in real time, adaptively adjust the sampling frequency, reduce redundant sampling of non-critical areas, and prioritize high-frequency sampling for fast-moving targets to capture details. Compared with traditional fixed sampling strategies, the sampling efficiency is improved by more than 40%. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0052] Figure 1 This is a flowchart of an adaptive sampling dynamic target tracking method based on single-pixel imaging, shown in an exemplary embodiment of the present application;
[0053] Figure 2 1 is a schematic diagram of the overall structure of a method for adaptive sampling dynamic target tracking based on single-pixel imaging, shown in an exemplary embodiment of the present application;
[0054] Figure 3 is a schematic diagram of a portion of a lighting pattern shown in an exemplary embodiment of the present application;
[0055] Figure 4 is a schematic diagram of another portion of an illumination pattern shown in an exemplary embodiment of the present application;
[0056] Figure 5It is a structural diagram of an adaptive sampling dynamic target tracking device based on single-pixel imaging shown in an exemplary embodiment of the present application.
[0057] Reference numerals:
[0058] 1. Photoelectric conversion unit; 2. Illumination pattern generation unit; 3. Projection unit; 4. Target scene; 5. Target object; 6. Illumination pattern; 7. Photoelectric detection unit; 8. Pattern generation module; 9. Signal modulation module; 10. Signal acquisition module; 11. Sampling frequency adjustment module; 12. Displacement calculation module. DETAILED DESCRIPTION
[0059] In order to make the purpose, implementation methods and advantages of the present application clearer, the exemplary implementation methods of the present application will be clearly and completely described below in conjunction with the drawings in the exemplary embodiments of the present application. Obviously, the described exemplary embodiments are only part of the embodiments of the present application, not all of the embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.
[0060] It should be noted that the brief descriptions of terms in this application are only for the purpose of facilitating the understanding of the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their ordinary and usual meanings.
[0061] In the specification and claims of this application and the accompanying drawings, the terms "first," "second," "third," etc. are used to distinguish similar or similar objects or entities, and are not necessarily intended to limit a particular order or sequence, unless otherwise noted. It should be understood that the terms used in this manner are interchangeable under appropriate circumstances.
[0062] The terms "comprise," "include," and "have," and any variations thereof, are intended to cover but not exclude inclusion; for example, a product or device comprising a list of components is not necessarily limited to all the components expressly listed but may include other components not expressly listed or inherent to such product or device.
[0063] Before explaining the adaptive sampling dynamic target tracking method based on single-pixel imaging provided in the embodiment of the present application, the application scenario and implementation environment of the embodiment of the present application are first introduced.
[0064] Object tracking is a key technology in fields such as computer vision, artificial intelligence, and robotics. It is especially crucial in applications that require real-time response and decision-making. In dynamic scenes, the target's motion, posture changes, and environmental complexity all pose huge challenges to target tracking.
[0065] Traditional target tracking methods often rely on multi-pixel sensors and complex computational models. While they can achieve good results in certain scenarios, they still have many shortcomings when faced with high-speed moving targets, complex backgrounds, and changing lighting conditions. Single-pixel imaging technology provides a new approach to dynamic target tracking. As an emerging computational imaging technology, it does not rely on traditional multi-pixel sensors and only requires a single light sensor for acquisition. It uses a single-pixel detector combined with structured light illumination and advanced image reconstruction algorithms to achieve target imaging and tracking. Its design has lower design costs and a simpler structure. Unlike traditional imaging technologies, single-pixel imaging technology uses a serial sampling method, which has stronger anti-interference capabilities in environments with poor lighting conditions. Due to its advantages such as wide operating spectrum, high sensitivity, and non-line-of-sight imaging, single-pixel imaging technology has unique advantages in infrared, ultraviolet, terahertz bands, as well as in imaging through scattering media and non-line-of-sight, making it suitable for target tracking tasks in a variety of complex environments.
[0066] Single-pixel computational imaging for tracking dynamic targets requires a computer to obtain an illumination pattern through a modulation algorithm. The illumination pattern is then projected onto the target scene through a spatial light modulator. The reflected light is used by a photodetector to obtain the light intensity signal value. Finally, the photoelectric signal is converted and the computer performs data acquisition and analysis to obtain the required information. The target's motion is analyzed without an image. The illumination patterns include randomly generated illumination patterns and digitally encoded illumination patterns. The reconstruction process of random illumination patterns is relatively cumbersome, computationally intensive, and has large reconstruction errors.
[0067] Therefore, the lighting pattern is generated in the form of digital coding, which greatly reduces the amount of calculation compared to randomly generated lighting patterns and improves the accuracy.
[0068] While the photodetector is collecting data, we use an Arduino development board to perform data acquisition and adaptive sampling. Adaptive sampling dynamically adjusts the sampling strategy by extracting important spectral coefficients during the data acquisition process to adaptively acquire spectral information. When the target is moving quickly, the sampling frequency is appropriately increased, while when the target is moving slowly, the sampling frequency is appropriately decreased. This method, based on real-time analysis of the image's spectral characteristics, effectively samples key locations. Research on adaptive sampling technology is gaining momentum in the field of single-pixel imaging, aiming to address the efficiency and quality limitations of traditional fixed sampling methods.
[0069] Based on this, this application provides a method and device for dynamic target tracking using adaptive sampling based on single-pixel imaging. This method uses a four-step phase-shift single-pixel imaging technique to generate a sine-cosine coded illumination pattern with different phase characteristics. This pattern is projected onto the target scene via a spatial light modulator. A photodetector receives the modulated reflected light signal and converts it into an electrical signal. The Fourier linear phase shift characteristic is then used to calculate the displacement of the dynamic target. An adaptive sampling strategy is employed to dynamically adjust the sampling frequency based on the target's motion speed: increasing the sampling rate at high speeds to capture details and decreasing it at low speeds to reduce redundant data.
[0070] Therefore, only a single-pixel imaging system is needed to achieve real-time tracking of fast-moving targets. Static interference is eliminated through the background frame difference method. Combined with eight circularly projected lighting patterns, the hardware complexity and cost are significantly reduced, data utilization and tracking accuracy are improved, and it is suitable for dynamic monitoring scenarios in complex environments.
[0071] Next, the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems will be described in detail through embodiments and in conjunction with the accompanying drawings. The various embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of them.
[0072] Figure 1 It is a flowchart of an adaptive sampling dynamic target tracking method based on single-pixel imaging shown in an exemplary embodiment of the present application.
[0073] In an exemplary embodiment, Figure 1 As shown, a method for adaptive sampling dynamic target tracking based on single-pixel imaging is provided. In this embodiment, the method may include the following steps:
[0074] Step 100: Generate four sets of sine-cosine coded illumination patterns with different phase characteristics.
[0075] Step 200: cyclically projecting the illumination pattern onto a target scene via a spatial light modulator to obtain a modulated signal.
[0076] Step 300: Receive the modulated reflected light signal of the target scene based on the photodetector, and convert the light signal into an electrical signal, which is received by the data acquisition card.
[0077] Step 400: Adopt an adaptive sampling strategy to dynamically adjust the sampling frequency, and control the maximum and minimum sampling times based on the voltage difference threshold.
[0078] Step 500: Calculate the target displacement using the Fourier linear phase shift characteristic, and combine it with the background frame difference method to eliminate static background interference and achieve real-time tracking of dynamic targets.
[0079] Figure 2 FIG. 1 is a schematic diagram of the overall structure of a method for adaptive sampling dynamic target tracking based on single-pixel imaging, shown in an exemplary embodiment of the present application. Figure 3 is a schematic diagram of a portion of a lighting pattern shown in an exemplary embodiment of the present application. Figure 4 FIG. 1 is a schematic diagram of another portion of an illumination pattern shown in an exemplary embodiment of the present application.
[0080] In one possible implementation, Figure 2 As shown, the specific implementation of the tracking method includes an illumination pattern generating unit 2, an illumination pattern 6 obtained by encoding 2, and the obtained pattern is projected into the target scene 4 by the projection unit 3. The modulated light intensity signal of the illumination pattern 6 and the target object 5 is collected by the photoelectric detection unit 7, and then the photoelectric conversion unit 1 converts the light signal into an electrical signal and adjusts the sampling rate. Finally, the data processing unit 2 calculates the displacement of the moving object to track and locate the target object 5.
[0081] More specifically, a computer is used to process and implement the lighting pattern generation unit and data processing, the photodetector is the main component of the photodetection unit, the photodiode is the main sensing device, and the photoelectric conversion unit is composed of an Arduino development board.
[0082] In simple terms, the whole process is that the lighting pattern generation unit 2 generates the lighting pattern 6 and gives the obtained pattern to the projection unit 3. The projection unit 3 projects the lighting pattern 6 into the target scene. The modulated reflected light is received by the photoelectric detection unit 7 and converted into an electrical signal by the photoelectric conversion unit 1. The current signal is collected using an adaptive sampling method, and the data processing unit 2 calculates the displacement, and finally monitors and tracks the target object 5.
[0083] The generation formula of the sine or cosine coding diagram with four phase characteristics is:
[0084] P1=m+ncos(2πf x x+2πf y y+0);
[0085]
[0086] P3=m+ncos(2πf x x+2πf y y+π);
[0087]
[0088] The lighting pattern here is generated using a Fourier basis, and can also be generated using a random matrix or a Hardmard matrix. The function of the lighting pattern generation unit 2 is to project the lighting pattern 6 onto the target scene. The lighting pattern generation unit 2 here can use a DLP projector, a spatial light modulator, and an LCD display. The purpose is only to project the lighting pattern into the scene. The light source in the scene can use natural light or externally generated light such as laser. The photoelectric conversion unit 1 converts the modulated light signal obtained by the photoelectric detection unit 7 into an electrical signal and stores and records it. Microcontrollers such as single-chip microcomputers, data acquisition cards, and FPGAs can also be used as photoelectric conversion units.
[0089] The lighting pattern generation unit 2 is also an important element for data processing. It further processes and analyzes the electrical signal obtained by the photoelectric conversion unit 1 to calculate the displacement of the target object and track the motion trajectory in real time.
[0090] Four coded patterns with different initial phases generated by a computer are projected into the target scene by a spatial light modulator. The photodetector receives the modulated reflected light signal and transmits it to the computer in the form of an electrical signal through photoelectric conversion to calculate the displacement of the target object.
[0091] The formula for calculating the displacement of the target object described above is:
[0092]
[0093] Among them, angle{} represents the operation of parameters, I represents the image of the target object, and I b represents the coefficients obtained from the background frame.
[0094] The background frame refers to a frame obtained before a moving object, here, a target object, enters the scene or starts to move.
[0095] By using the background elimination method, the current single-pixel measurement value and the voltage difference when the object enters the scene or starts to move out of the scene are obtained to judge and analyze the movement of the target object.
[0096] The above-mentioned four-step phase-shift single-pixel imaging method is characterized in that: the required projection image is composed of f x =2 / 128, f y = 0 and the four illumination patterns with different phases and f x =0,f y =2 / 128, four illumination patterns with different phases are obtained, and a total of eight patterns are projected cyclically.
[0097] The above-mentioned target object movement tracking is characterized in that: no complex monitoring equipment is required, and only a single-pixel imaging system is required to track and locate fast-moving objects in real time.
[0098] The number of projection images required is the two sets of four illumination patterns with different phases as described above, a total of eight images, and is determined by the specified number of cycles, that is, 8 times the number of cycles.
[0099] like Figure 3 and Figure 4 As shown, the illumination pattern is a schematic diagram of 8 sine or cosine coding diagrams with four phase characteristics.
[0100] Some embodiments of the present application innovatively achieve efficient generation of illumination patterns by using four sine or cosine coding images with different phases on the x and y axes, respectively.
[0101] Compared with traditional single-pixel technology, this method significantly reduces the number of required illumination patterns and is superior to other image computational imaging methods in terms of accuracy. In addition, it has low computational complexity and can complete the displacement calculation of dynamic targets in a shorter time. It demonstrates unparalleled performance in real-time applications and large-scale image processing. Moreover, it can achieve accurate tracking and positioning of target objects without the need for excessive target tracking tools.
[0102] It should be understood that, although the various steps in the flowcharts involved in the above-described embodiments are displayed in sequence according to the instructions, these steps are not necessarily executed in the order indicated. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.
[0103] Corresponding to the aforementioned embodiment of the adaptive sampling dynamic target tracking method based on single-pixel imaging, adopting the same technical concept, the present application also provides an embodiment of the adaptive sampling dynamic target tracking device based on single-pixel imaging.
[0104] Figure 5 It is a structural diagram of an adaptive sampling dynamic target tracking device based on single-pixel imaging shown in an exemplary embodiment of the present application.
[0105] In an exemplary embodiment, Figure 5As shown, the adaptive sampling dynamic target tracking device based on single-pixel imaging includes:
[0106] a pattern generating module 8, configured to generate four sets of sine-cosine coded lighting patterns with different phase characteristics;
[0107] a signal modulation module 9, configured to cyclically project the illumination pattern onto a target scene via a spatial light modulator to obtain a modulation signal;
[0108] The signal acquisition module 10 is used to receive the modulated reflected light signal of the target scene based on the photoelectric detector and convert the light signal into an electrical signal, which is received by the data acquisition card;
[0109] The sampling frequency adjustment module 11 is used to dynamically adjust the sampling frequency using an adaptive sampling strategy and control the maximum and minimum sampling times based on the voltage difference threshold;
[0110] The displacement calculation module 12 is used to calculate the target displacement using the Fourier linear phase shift characteristic, and to eliminate static background interference in combination with the background frame difference method to achieve real-time tracking of dynamic targets.
[0111] For the specific limitations of the adaptive sampling dynamic target tracking device based on single-pixel imaging, please refer to the limitations of the adaptive sampling dynamic target tracking method based on single-pixel imaging above, which will not be repeated here. Each module in the above-mentioned adaptive sampling dynamic target tracking device based on single-pixel imaging can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0112] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0113] The embodiments described above merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.
Claims
1. A method for dynamic target tracking based on adaptive sampling of single-pixel imaging, characterized in that: include: Generate four sets of sine-cosine coded illumination patterns with different phase characteristics; cyclically projecting the illumination pattern onto a target scene through a spatial light modulator to obtain a modulated signal; The photoelectric detector receives the modulated reflected light signal of the target scene and converts the light signal into an electrical signal, which is received by the data acquisition card; Adopting adaptive sampling strategy to dynamically adjust the sampling frequency, and controlling the maximum and minimum sampling times based on the voltage difference threshold; The target displacement is calculated using the Fourier linear phase shift characteristic, and the background frame difference method is combined to eliminate static background interference and achieve real-time tracking of dynamic targets.
2. The method for dynamic target tracking based on adaptive sampling of single-pixel imaging according to claim 1, wherein: The generation formula of the lighting pattern includes: P1=m+ncos(2πf x x+2πf y y+0); P3=m+ncos(2πf x x+2πf y y+π); Among them, m is the average light intensity, n is the contrast, x and y are the coordinates of the plane where the target object is located, and f x and f y are the spatial frequencies in the x and y directions respectively.
3. The method for dynamic target tracking based on adaptive sampling of single-pixel imaging according to claim 2, wherein: The illumination pattern includes two sets of spatial frequency combinations including: Group 1: f x =2 / 128, f y =0; Group 2: f x =0,f y =2 / 128; Each group contains four phases with different phases. A total of eight lighting patterns are projected in a loop.
4. The method for dynamic target tracking based on adaptive sampling of single pixel imaging according to claim 3, wherein: The rules of the adaptive sampling strategy are: Where |Vi-Vi-1| represents the voltage difference between the i-th sample and the i-1-th sample, T is the voltage change threshold, and Nimax and Nimin are the dynamically adjusted maximum and minimum sampling times, respectively. These values increase or decrease based on voltage fluctuations, controlling the sampling range and ensuring increased sampling during rapid changes and decreased sampling during slow changes, thereby balancing resource usage and accuracy. Sfast and Sslow are the sample count adjustment steps for fast and slow sampling, respectively.
5. The method for dynamic target tracking based on adaptive sampling of single pixel imaging according to claim 4, characterized in that: The adaptive sampling strategy is adopted when collecting data, increasing the sampling rate when the target object moves faster and reducing the sampling rate when the target object moves slower, so as to achieve a balance between data utilization and system, thereby reducing unnecessary data redundancy.
6. The method for dynamic target tracking based on adaptive sampling of single pixel imaging according to claim 1, wherein: The calculation formula of the target displacement is: Among them, angle{} represents the operation of parameters, I represents the image of the target object, and I b Represents the coefficients obtained from the background frame, which is the frame obtained before the target object enters the scene or starts moving. x0 and y0 are the displacements calculated in the x and y directions, respectively.
7. The method for dynamic target tracking based on adaptive sampling of single pixel imaging according to claim 1, wherein: The formula for obtaining the target scene is: F = (D1-D3) + j·(D2-D4); I(x-x0,y-y0)=F -1 {I(f x ,f y )exp[-j2π(f x x0+f x y0)]}; Among them, D1, D2, D3, and D4 are the electrical signals corresponding to P1, P2, P3, and P4 in the above formula, respectively, and are also the electrical signals converted by the photoelectric unit. F represents the result of Fourier transform, which converts the image from the spatial domain to the frequency domain through a four-step phase shift method. j represents the sign of the imaginary part.
8. The method for dynamic target tracking based on adaptive sampling of single pixel imaging according to claim 1, wherein: The spatial light modulator is a DLP projector or an LCD display screen, which is used to project the sine-cosine coded illumination pattern. The photodetector is a photodiode, which receives the reflected light signal and outputs an electrical signal through a photoelectric conversion unit.
9. The method for dynamic target tracking based on adaptive sampling of single pixel imaging according to claim 1, wherein: Also includes: The reconstructed image is projected back into the scene via a spatial light modulator, enabling intuitive presentation of the target object's position information. The changes in the pixels in the final reconstructed pattern directly reflect the location information of the target object; By comparing the calculation results of different pixel points, the specific position of the target object in the scene can be determined quickly and accurately.
10. An adaptive sampling dynamic target tracking device based on single-pixel imaging, characterized in that: include: a pattern generation module, configured to generate four sets of sine-cosine coded illumination patterns with different phase characteristics; a signal modulation module, configured to cyclically project the illumination pattern onto a target scene via a spatial light modulator to obtain a modulated signal; A signal acquisition module is used to receive the modulated reflected light signal of the target scene based on a photoelectric detector and convert the light signal into an electrical signal to be received by a data acquisition card; Sampling frequency adjustment module, used to dynamically adjust the sampling frequency using an adaptive sampling strategy and control the maximum and minimum sampling times based on the voltage difference threshold; The displacement calculation module is used to calculate the target displacement using the Fourier linear phase shift characteristics, and combine the background frame difference method to eliminate static background interference and realize real-time tracking of dynamic targets.