A single-frame particle image motion vector measurement method based on time encoding illumination

Abstract

The application discloses a single-frame particle image motion vector measurement method based on time coding illumination, and relates to the technical field of particle motion measurement. The method comprises the following steps: S1, during single exposure of an image sensor, pre-defined time sequence modulated illumination is used to irradiate tracer particles in a flowing medium, so that multiple continuous motion trajectories of the tracer particles are formed in a single frame image, and time marker points generated by pulsed illumination are embedded on each trajectory; S2, spatial information of each tracer particle motion trajectory in a field of view is identified and extracted from the single frame image; and S3, for each extracted trajectory, motion vector information corresponding to the trajectory is calculated based on the geometric shape of the trajectory, the spatial position of the illumination marker points and pre-defined non-uniform time intervals. The application improves the time resolution and measurement accuracy, and can meet the low-cost, high-precision and high-efficiency particle motion measurement requirements in various transient and steady state scenes.

Description

Technical Field

[0001] This invention relates to the field of particle motion measurement technology, and in particular to a method for measuring the motion vector of a single-frame particle image based on time-coded illumination. Background Technology

[0002] In fields such as flow field analysis and biomedical imaging, accurately acquiring the velocity and direction of particle motion is a core requirement for analyzing dynamic processes. Particle image velocimetry (PIV) technology, as a key method, has developed into two main technical approaches: Particle Image Velocimetry (PIV) and Particle Tracking Velocimetry (PTV). PIV technology indirectly estimates the velocity distribution of the flow field by calculating the cross-correlation function of a particle swarm in two consecutive frames. PTV technology, on the other hand, locates the discrete positions of individual particles across multiple frames and then connects these discrete points using a trajectory association algorithm to obtain the particle's motion vector.

[0003] However, existing technologies are limited by the underlying paradigm of multi-frame discrete sampling, and generally regard motion blur generated during particle motion as harmful noise that needs to be suppressed, resulting in difficult-to-overcome defects in practical applications: High frequency hardware dependence: In order to satisfy the sampling theorem and avoid calculation errors caused by excessive inter-frame particle displacement, existing technologies must use kHz-level high-speed cameras as the core of image acquisition. Such devices are not only expensive to purchase, but also require high-bandwidth data storage and transmission modules, which greatly increases the deployment threshold of the system and makes it difficult to adapt to miniaturized and portable application scenarios. Low temporal resolution: Whether it is the two-frame cross-correlation calculation of PIV or the multi-frame trajectory stitching of PTV, at least two frames of image data are required to generate a complete velocity field result. Its actual temporal resolution is far lower than the camera's nominal frame rate. It cannot effectively capture and analyze transient dynamic processes such as intravascular blood flow pulsation and sudden fluid turning in microfluidic chips. Limited spatial accuracy: PTV technology constructs motion trajectories by connecting the discrete positioning points of particles in adjacent frames with straight lines. This discrete point-straight line fitting method cannot restore the true continuous motion path of particles. Especially in areas with large flow field streamline curvature, the trajectory geometric distortion phenomenon is significant, which directly affects the accuracy of velocity calculation. Poor real-time performance: Existing technologies require computational steps such as multi-frame image registration, particle recognition and matching, and complex trajectory association. Not only is the algorithm logic cumbersome, but it also places high demands on hardware computing power, making it difficult to achieve real-time processing and analysis of flow field data. At the same time, the algorithm performance is highly sensitive to parameters such as registration accuracy and particle density threshold. Fine-tuning of parameters may lead to deviations in results, resulting in insufficient stability and robustness.

[0004] In summary, existing technologies have failed to break through the traditional framework of suppressing motion blur and relying on multi-frame sampling, and cannot fundamentally solve the contradiction between hardware cost, temporal resolution, trajectory fidelity and processing efficiency, thus restricting the application expansion of particle image velocimetry technology in low-cost, high-dynamic, and high-real-time scenarios. Summary of the Invention

[0005] The purpose of this invention is to propose a single-frame particle image motion vector measurement method based on time-coded illumination. The complete measurement of particle motion vectors can be completed through single-frame imaging, without relying on multi-frame image sequences, which reduces hardware dependence, improves temporal resolution, avoids trajectory fitting distortion, and simplifies the data processing flow.

[0006] To achieve the above objectives, this invention proposes a method for measuring motion vectors in a single-frame particle image based on time-coded illumination, the specific steps of which are as follows: Step S1: Time-coded image acquisition. During a single exposure of the image sensor, tracer particles in the flowing medium are illuminated with predefined time-modulated illumination light, so that multiple continuous motion trajectories of tracer particles are formed in a single frame image, and time markers generated by pulse illumination are embedded on each trajectory. Step S2: Trajectory information extraction. Identify and extract the spatial information of each tracer particle motion trajectory in the field of view from a single frame image, including the geometry of the trajectory and the spatial position of the illumination marker point on the trajectory. Step S3: Vector information decoding. For each extracted trajectory, based on the trajectory's geometry, the spatial position of the lighting markers, and a predefined non-uniform time interval, the motion vector information corresponding to the trajectory is calculated. The motion vector information includes velocity and direction.

[0007] Preferably, the specific implementation steps for acquiring the time-coded image in step S1 are as follows: Step S111: Setting up the illumination system: Using a modulated light source to ensure that the illumination light emits high-intensity pulses in a predefined sequence during exposure; the illumination system includes a dual-laser scheme and a single-laser dual-beam scheme. The dual-laser scheme uses a continuous-wave laser to provide substrate illumination and a pulsed laser to provide coded pulses; the single-laser dual-beam scheme uses the same laser output split into two paths, one as a continuous beam to provide substrate illumination, and the other as a high-intensity pulse after modulation by a modulator; the modulator type is an acousto-optic modulator or an electro-optic modulator. Step S112, Synchronization Control: The camera exposure trigger signal and illumination pulse trigger signal are generated simultaneously by an external multi-channel synchronization controller to ensure that the predefined pulse time point falls accurately within the camera's single exposure time window; Step S113, Exposure and Acquisition: The camera performs a single exposure to record the trajectory of the tracer particles in the flowing medium; during the exposure, the illumination system triggers no fewer than two high-intensity pulses at non-uniform time intervals to form illumination markers on the trajectory.

[0008] Preferably, in step S113, the non-uniform time interval satisfies: Let the single exposure period of the image sensor be... ,in The moment the exposure begins. This is the end time of the exposure; during a single exposure, it is triggered according to a predefined timing sequence. n A high-intensity lighting pulse, the pulse triggering time sequence is as follows: ,satisfy The time interval between any two adjacent pulses is ,in, Time interval The set of time intervals formed There exists at least one pair of intervals with unequal lengths, such that .

[0009] Preferably, to improve the measurement accuracy of particle motion vectors at different velocities and enhance the robustness of direction determination, the pulse triggering time sequence is... During the exposure cycle The distribution within the system is optimized.

[0010] Preferably, in step S1, each high-intensity pulse generates a local high-brightness region on the trajectory of the tracer particle, serving as a time marker. This time marker is distinguished from the base illumination portion of the trajectory by one or more predefined physical features, including intensity features, shape features, and spectral features. The intensity feature is that the signal intensity of the marker region is higher than that of the continuous trajectory segment. The shape feature is that the pulse spot presents a specific and easily identifiable shape on the imaging plane through optical design or modulation. The spectral feature is that the pulse is generated using a light source with a different wavelength than the base illumination, creating a spectral difference.

[0011] Preferably, step S1 also includes a system parameter design process, the specific steps of which are as follows: Step S121: Determine the minimum analysis trajectory arc length and the basic exposure time; where, the minimum analysis trajectory arc length... To accommodate multiple marker points and enable reliable analysis; based on and minimum expected speed Calculate the base exposure time At the same time, a safety factor is introduced. Determine the minimum exposure time ; Step S122: Based on the camera frame rate system constraints, select the actual exposure time used by the system. ,satisfy Set the exposure period to ,in ; Step S123: Calculate the theoretical minimum margin Verify the minimum physical time resolution achievable by the system hardware. ,satisfy ;in, For system spatial resolution, For the maximum expected speed; Step S124: Select the number of pulses , ; Step S125, Set pulse position: In Internally set pulse trigger time ,satisfy ; Step S126: Calculate the time interval between all adjacent pulses. ,in And ensure that at least one pair exists. ; Step S127: Verify the spatial resolvability of the marker points, satisfying: ;in, For system spatial resolution, As the slowest particle, The shortest pulse time interval; Step S128: Verify the set longest pulse time interval. satisfy ,in, For the field of view, The fastest particle; Step S129: Iterative optimization. If steps S125-S128 do not meet the constraints, return to adjust the number of pulses. Pulse sequence or exposure time Re-evaluate system performance metrics until all constraints are met.

[0012] Preferably, in step S2, the specific process of trajectory information extraction is as follows: preprocessing operations such as background correction, noise suppression, contrast enhancement, and structure enhancement are performed on the original image to enhance the signal-to-noise ratio and trajectory features; candidate trajectory regions are separated from the preprocessed image and trajectory geometric information is extracted; finally, based on spatial location distribution features and predefined pulse timing, illumination markers are located on the extracted trajectory and paired and sorted.

[0013] Preferably, specific methods for image preprocessing include: Background correction is performed using particle-free background image subtraction, dynamic background modeling, or high-frequency filtering. Noise suppression can be achieved using spatial domain filtering or frequency domain filtering. Contrast enhancement is achieved by employing global or adaptive histogram equalization, contrast-limited adaptive histogram equalization, or image enhancement based on grayscale transformation functions. Structural enhancement is achieved using filters or morphological operations that enhance linear structures.

[0014] Preferably, in step S3, the specific implementation steps for vector information decoding are as follows: Step S31: Calculate particle velocity: Based on the trajectory arc length, the arc length of all or part of the actual motion path is obtained by integrating the trajectory centerline, and the speed is calculated in combination with the corresponding time interval. For the complete trajectory, the magnitude of the velocity is equal to the total arc length. With exposure time The quotient, the formula is: ;in, The particle velocity; For a portion of the trajectory arc length, the particle's velocity is the quotient of the corresponding arc length and the time interval, as shown in the formula: ;in, Mark the point and The velocity of the particle over the arc length, the time interval , , Representing the marker points and The corresponding time point, Mark the point and The arc length between; Step S32, determining the direction of particle motion, is as follows: Compare the temporal order of the marker points with the spatial order along the arc length of the trajectory. If the direction of increasing arc length is consistent with the direction of time progression, then the flow direction is... Otherwise ;in, As the starting point of the particle, The endpoint of the particle; By calculating the arc length ratio of adjacent segments... Compared with the known time interval : like The flow direction is ; like The flow direction is ; in, , ; , These represent the arc lengths of two adjacent segments on the trajectory. , These represent the time intervals between two adjacent trajectory segments. For the marked point, and .

[0015] Therefore, this invention proposes a method for measuring motion vectors in a single-frame particle image based on time-coded illumination, which has the following advantages: (1) The single-frame particle image motion vector measurement method based on time-coded illumination proposed in this invention does not rely on multi-frame image sequences, which significantly reduces the dependence on high-speed hardware and reduces the measurement cost; (2) The time resolution of the single-frame particle image motion vector measurement method based on time-coded illumination proposed in this invention is greatly improved to the theoretical limit of camera frame rate, which can meet the high-precision measurement requirements of transient flow and other scenarios. (3) The single-frame particle image motion vector measurement method based on time-coded illumination proposed in this invention effectively avoids trajectory fitting distortion in traditional multi-frame tracking through continuous trajectory analysis, thereby improving measurement accuracy.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] Figure 1 This is a flowchart of a single-frame particle image motion vector measurement method based on time-coded illumination according to the present invention; Figure 2 This is a demonstration diagram of a single-frame particle image motion vector measurement method based on time-coded illumination according to the present invention. Detailed Implementation

[0018] To make the technical solutions, advantages, and objectives of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below. The described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0020] Example like Figures 1-2 As shown, this invention provides a method for measuring motion vectors in a single-frame particle image based on time-coded illumination, the steps of which are as follows: Step S1: Time-coded image acquisition. During a single exposure of the image sensor, tracer particles in the flowing medium are illuminated with predefined time-modulated illumination light, so that multiple continuous motion trajectories of tracer particles are formed in a single frame image, and time markers generated by pulse illumination are embedded on each trajectory. The specific implementation steps for acquiring time-coded images are as follows: Step S111: Setting up the illumination system: Using a modulated light source to ensure that the illumination light emits high-intensity pulses in a predefined sequence during exposure; the illumination system includes a dual-laser scheme and a single-laser dual-beam scheme. The dual-laser scheme uses a continuous-wave laser to provide substrate illumination and a pulsed laser to provide coded pulses; the single-laser dual-beam scheme uses the same laser output split into two paths, one as a continuous beam to provide substrate illumination, and the other as a high-intensity pulse after modulation by a modulator; the modulator type is an acousto-optic modulator or an electro-optic modulator. Step S112, Synchronization Control: The camera exposure trigger signal and illumination pulse trigger signal are generated simultaneously by an external multi-channel synchronization controller to ensure that the predefined pulse time point falls accurately within the camera's single exposure time window; Step S113, Exposure and Acquisition: The camera performs a single exposure to record the trajectory of tracer particles in the flowing medium; during the exposure, the illumination system triggers at least two high-intensity pulses at non-uniform time intervals, forming illumination markers on the trajectory; wherein the non-uniform time intervals satisfy: Let the single exposure period of the image sensor be... ,in The moment the exposure begins. This is the end time of the exposure; during a single exposure, it is triggered according to a predefined timing sequence. n A high-intensity lighting pulse, the pulse triggering time sequence is as follows: ,satisfy The time interval between any two adjacent pulses is ,in, Time interval The set of time intervals formed There exists at least one pair of intervals with unequal lengths, such that .

[0021] To improve the measurement accuracy of particle motion vectors at different velocities and enhance the robustness of direction determination, the pulse trigger time sequence was analyzed. During the exposure cycle The distribution within the system is optimized, and the specific optimization principles include: Avoid concentrated distribution; avoid excessive concentration of pulse timing at the beginning or end of the exposure cycle. The goal is to achieve a near-uniform division, ensuring that the trajectory of the pulse moment can be divided approximately uniformly in time while satisfying the non-uniformity constraint.

[0022] The specific implementation steps for acquiring time-coded images also include a system parameter design process, the specific steps of which are as follows: Step S121: Determine the minimum analysis trajectory arc length and the basic exposure time; where, the minimum analysis trajectory arc length... To accommodate multiple marker points and enable reliable analysis; based on and minimum expected speed Calculate the base exposure time At the same time, a safety factor is introduced. Determine the minimum exposure time ; Step S122: Based on the camera frame rate system constraints, select the actual exposure time used by the system. ,satisfy Set the exposure period to ,in ; Step S123: Calculate the theoretical minimum margin Verify the minimum physical time resolution achievable by the system hardware. ,satisfy ;in, For system spatial resolution, For the maximum expected speed; Step S124: Select the number of pulses , ; Step S125, Set pulse position: In Internally set pulse trigger time ,satisfy ; Step S126: Calculate the time interval between all adjacent pulses. ,in And ensure that at least one pair exists. ; Step S127: Verify the spatial resolvability of the marker points, satisfying: ;in, For system spatial resolution, As the slowest particle, The shortest pulse time interval; Step S128: Verify the set longest pulse time interval. satisfy ,in, For the field of view, The fastest particle; Step S129: Iterative optimization. If steps S125-S128 do not meet the constraints, return to adjust the number of pulses. Pulse sequence or exposure time Re-evaluate system performance metrics until all constraints are met.

[0023] Each high-intensity pulse generates a localized high-brightness region on the trajectory of the tracer particle, serving as a time marker. This time marker is distinguished from the substrate illumination portion of the trajectory by one or more predefined physical features. These physical features include intensity, shape, and spectral characteristics: the intensity feature is that the signal intensity in the marker region is higher than that in the continuous trajectory segment; the shape feature is that the pulse spot presents a specific and easily identifiable shape on the imaging plane through optical design or modulation; and the spectral feature is that the pulse is generated using a light source with a different wavelength than the substrate illumination, creating a spectral difference.

[0024] Step S2, Trajectory Information Extraction: This step identifies and extracts the motion of each tracer particle within the field of view from a single-frame image, including the geometry of the trajectory and the spatial position of the illumination markers on the trajectory. Specifically, the process involves: preprocessing the original image with background correction, noise suppression, contrast enhancement, and structural enhancement to improve the signal-to-noise ratio and trajectory features; separating candidate trajectory regions from the preprocessed image and extracting trajectory geometric information; and finally, based on spatial distribution features and a predefined pulse timing sequence, locating and pairing illumination markers on the extracted trajectory. The specific image preprocessing methods include: Background correction is performed using particle-free background image subtraction, dynamic background modeling, or high-frequency filtering. Noise suppression can be achieved using spatial domain filtering or frequency domain filtering. Contrast enhancement is achieved by employing global or adaptive histogram equalization, contrast-limited adaptive histogram equalization, or image enhancement based on grayscale transformation functions. Structural enhancement is achieved using filters or morphological operations that enhance linear structures.

[0025] Step S3: Vector Information Decoding. For each extracted trajectory, based on the trajectory's geometry, the spatial position of the lighting markers, and a predefined non-uniform time interval, the motion vector information corresponding to the trajectory is calculated. The motion vector information includes velocity and direction. The specific implementation steps are as follows: Step S31: Calculate particle velocity: Based on the trajectory arc length, the arc length of all or part of the actual motion path is obtained by integrating the trajectory centerline, and the speed is calculated in combination with the corresponding time interval. For the complete trajectory, the magnitude of the velocity is equal to the total arc length. With exposure time The quotient, the formula is: ;in, The particle velocity; For a portion of the trajectory arc length, the particle's velocity is the quotient of the corresponding arc length and the time interval, as shown in the formula: ;in, Mark the point and The velocity of the particle over the arc length, the time interval , , Representing the marker points and The corresponding time point, Mark the point and The arc length between; Step S32, determining the direction of particle motion, is as follows: Compare the temporal order of the marker points with the spatial order along the arc length of the trajectory. If the direction of increasing arc length is consistent with the direction of time progression, then the flow direction is... Otherwise ;in, As the starting point of the particle, The endpoint of the particle; By calculating the arc length ratio of adjacent segments... Compared with the known time interval : like The flow direction is ; like The flow direction is ; in, , ; , These represent the arc lengths of two adjacent segments on the trajectory. , These represent the time intervals between two adjacent trajectory segments. For the marked point, and .

[0026] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.

[0027] Therefore, this invention provides a single-frame particle image motion vector measurement method based on time-coded illumination. The particle motion vector parameters can be decoded through single-frame imaging, which reduces the dependence on high-speed hardware, improves the temporal resolution and measurement accuracy, simplifies the data processing process, and can meet the needs of low-cost, high-precision, and high-efficiency particle motion measurement in various transient and steady-state scenarios.

[0028] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for measuring motion vectors in a single-frame particle image based on time-coded illumination, characterized in that, The specific steps are as follows: Step S1: Time-coded image acquisition. During a single exposure of the image sensor, tracer particles in the flowing medium are illuminated with predefined time-modulated illumination light, so that multiple continuous motion trajectories of tracer particles are formed in a single frame image, and time markers generated by pulse illumination are embedded on each trajectory. Step S2: Trajectory information extraction. Identify and extract the spatial information of each tracer particle motion trajectory in the field of view from a single frame image, including the geometry of the trajectory and the spatial position of the illumination marker point on the trajectory. Step S3: Vector information decoding. For each extracted trajectory, based on the trajectory's geometry, the spatial position of the lighting markers, and a predefined non-uniform time interval, the motion vector information corresponding to the trajectory is calculated. The motion vector information includes velocity and direction. In step S1, the specific implementation steps for acquiring the time-coded image are as follows: Step S111: Set up the lighting system: Use a modulated light source to ensure that the lighting energy emits high-intensity pulses in a predefined sequence during exposure; Step S112, Synchronization Control: The camera exposure trigger signal and illumination pulse trigger signal are generated simultaneously by an external multi-channel synchronization controller to ensure that the predefined pulse time point falls accurately within the camera's single exposure time window; Step S113, Exposure and Acquisition: The camera performs a single exposure to record the trajectory of the tracer particles in the flowing medium; during the exposure, the illumination system triggers no fewer than two high-intensity pulses at non-uniform time intervals to form illumination markers on the trajectory; The lighting system includes a dual-laser scheme and a single-laser dual-beam scheme. The dual-laser scheme uses a continuous-wave laser to provide base illumination and a pulsed laser to provide coded pulses. The single-laser dual-beam scheme uses a single laser whose output is split into two paths: one is a continuous beam to provide base illumination, and the other is modulated by a modulator to generate high-intensity pulses. The modulator type is an acousto-optic modulator or an electro-optic modulator. In step S1, each high-intensity pulse generates a local high-brightness region on the trajectory of the tracer particle as a time marker. The time marker is distinguished from the base illumination portion of the trajectory by one or more predefined physical features, including intensity features, shape features, and spectral features.

2. The method for measuring motion vectors of a single-frame particle image based on time-coded illumination according to claim 1, characterized in that, In step S113, the non-uniform time interval satisfies: Let the single exposure period of the image sensor be... ,in The moment the exposure begins. This is the end time of the exposure; during a single exposure, it is triggered according to a predefined timing sequence. n A high-intensity lighting pulse, the pulse triggering time sequence is as follows: ,satisfy ; The time interval between any two adjacent pulses is ,in, Time interval The set of time intervals formed There exists at least one pair of intervals with unequal lengths, such that .

3. The method for measuring motion vectors of a single-frame particle image based on time-coded illumination according to claim 2, characterized in that, To improve the measurement accuracy of particle motion vectors at different velocities and enhance the robustness of direction determination, the pulse trigger time sequence was analyzed. During the exposure cycle The distribution within the system is optimized.

4. The method for measuring motion vectors of a single-frame particle image based on time-coded illumination according to claim 3, characterized in that, The intensity feature is that the signal intensity in the marked point region is higher than that in the continuous trajectory segment; the shape feature is that the pulse spot presents a specific and easily identifiable shape on the imaging plane through optical design or modulation; the spectral feature is that the pulse is generated by using a light source with a different wavelength from the substrate illumination to form a spectral difference.

5. The method for measuring motion vectors of a single-frame particle image based on time-coded illumination according to claim 4, characterized in that, In step S2, the specific process of trajectory information extraction is as follows: preprocessing operations such as background correction, noise suppression, contrast enhancement, and structure enhancement are performed on the original image to enhance the signal-to-noise ratio and trajectory features; Candidate trajectory regions are separated from the preprocessed image and trajectory geometric information is extracted. Finally, based on spatial location distribution features and predefined pulse timing, illumination markers are located on the extracted trajectories and paired and sorted.

6. The method for measuring motion vectors of a single-frame particle image based on time-coded illumination according to claim 5, characterized in that, Specific methods for image preprocessing include: Background correction is performed using particle-free background image subtraction, dynamic background modeling, or high-frequency filtering. Noise suppression can be achieved using spatial domain filtering or frequency domain filtering. Contrast enhancement is achieved by employing global or adaptive histogram equalization, contrast-limited adaptive histogram equalization, or image enhancement based on grayscale transformation functions. Structural enhancement is achieved using filters or morphological operations that enhance linear structures.

Images