Three-dimensional imaging and detection integrated device and method for carbon deuterium pellets

Abstract

The invention discloses a three-dimensional imaging detection integrated device and method for a carbon deuterium pellet, and belongs to the technical field of optical precision measurement and nondestructive testing. The device comprises an illumination mechanism, a light source mechanism and a control mechanism, wherein the illumination mechanism adopts an addressable LED array to provide multi-angle illumination; the imaging mechanism is a 4f microscopic system and is used for collecting images; the sample control mechanism is used for realizing automatic posture adjustment and transmission of the pellets; and the control mechanism coordinates all the modules to complete automatic detection. The method comprises the steps that an image sequence is collected, complex amplitude projection is recovered through a Fourier lamination imaging algorithm, and based on a tomography inversion algorithm, the surface point cloud radius is calculated and compared with a standard value, and the defect is automatically identified and marked. According to the invention, automatic, quantitative and high-resolution three-dimensional imaging and evaluation of the morphology and internal defects of the pellet are realized, and a stable and efficient measurement platform is provided for scientific research and industrial application.

Description

Technical Field

[0001] This invention relates to the field of optical precision measurement and non-destructive testing technology, specifically to an integrated three-dimensional imaging detection device and method for carbon-deuterium targets. Background Technology

[0002] In inertial confinement fusion (ICF) research, the target pellet, as the carrier of fusion fuel, requires strict control over its geometric precision, shell homogeneity, and internal defects to achieve symmetrical implosion and successful ignition. Any micron-level morphological defects or structural inhomogeneities can be drastically amplified under Rayleigh-Taylor instability, leading to experimental failure. Therefore, developing technologies and equipment capable of high-precision, non-destructive, and quantitative three-dimensional imaging and detection of the target pellet has become a crucial aspect of ICF research.

[0003] Currently, detection methods for such transparent three-dimensional microsphere samples have significant limitations. While X-ray computational computed tomography (CT) possesses strong penetrating power, its imaging contrast depends on the material's absorption coefficient. For targets composed of light elements such as carbon and deuterium, the differences in X-ray absorption are weak, resulting in insufficient resolution of internal fine structures and difficulty in accurately characterizing refractive index distribution. On the other hand, traditional optical microscopy techniques, such as interferometric microscopy, while providing high phase sensitivity in the visible light band, often require the use of refractive index matching fluids to eliminate the influence of refractive interfaces when dealing with samples with curved structures. This process is cumbersome and may introduce errors or contaminate the sample. Furthermore, most existing detection equipment has limited functions or relies on manual operation for multi-angle data acquisition and alignment. It suffers from significant deficiencies in the system integration of optics, mechanics, electronics, and control, automated detection processes, and overall imaging stability, failing to meet the urgent needs of scientific research and engineering applications for high-throughput, repeatable, and one-click three-dimensional imaging. Although existing research has attempted to apply technologies such as machine vision and optical tweezers manipulation to microsphere detection, these solutions have mostly remained at the stage of principle verification or modularization, failing to form a complete solution that integrates automatic lighting control, precise motion coordination, rapid image acquisition, robust 3D reconstruction, and automatic defect analysis.

[0004] Therefore, existing technologies (such as X-ray CT and traditional optical microscopy) have shortcomings when detecting transparent three-dimensional microspheres such as ICF targets, including low imaging contrast, reliance on manual operation and refractive index matching fluid, and insufficient system automation integration, which makes it impossible to achieve high-throughput, high-precision automated three-dimensional non-destructive testing. Summary of the Invention

[0005] The purpose of this invention is to provide an integrated device and method for three-dimensional imaging and detection of carbon-deuterium targets, so as to overcome the above-mentioned problems in the prior art and realize automatic, quantitative, high-resolution three-dimensional imaging and evaluation of the morphology and internal defects of the targets.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A three-dimensional imaging and detection integrated device for carbon-deuterium targets, comprising: Lighting mechanism, including addressable LED array light source and its driving circuitry; An imaging mechanism, optically connected to the illumination mechanism, includes a microscopic optical system and a camera, wherein the microscopic optical system consists of objective lenses and telescopes arranged in a 4f optical path structure; The sample handling mechanism includes: The adsorption rotation unit, consisting of a rotary motor and a transparent suction nozzle mechanically connected, is used to fix and rotate the target pellet. The displacement positioning unit, consisting of a three-dimensional displacement stage and a rotary stage mechanically connected, is used to control the spatial position of the adsorption rotation unit. The cargo unit is layered and has built-in tracks for storing targets to be tested and those already tested. The control mechanism is electrically connected to the lighting mechanism, imaging mechanism, and sample manipulation mechanism, and includes a main controller and a slave controller that communicates with the main controller. The output of the slave controller is connected to the driving circuit of the LED array light source and the rotary motor, respectively. The main controller is also communicatively connected to the displacement stage controller and the camera.

[0007] Furthermore, in the adsorption rotation unit, the central axis of the transparent suction nozzle is coaxially arranged with the rotation axis of the rotary motor.

[0008] Furthermore, the carrying unit includes a layered chamber for testing and a chamber that has already been tested, and an internal track is provided for guiding the target pellet to move between the chamber for testing and the chamber that has already been tested.

[0009] Furthermore, in the 4f optical path structure, the distance from the sample to the front focal plane of the objective lens is the focal length of the objective lens, the distance from the rear focal plane of the objective lens to the front focal plane of the simplified lens is the sum of the focal lengths of the objective lens and the simplified lens, and the distance from the rear focal plane of the simplified lens to the target surface of the camera is the focal length of the simplified lens.

[0010] Furthermore, the controller includes a microcontroller, which is an Arduino microcontroller.

[0011] Another object of the present invention is to provide a three-dimensional imaging detection method for carbon-deuterium targets, the method employing the aforementioned integrated three-dimensional imaging detection device for carbon-deuterium targets, and comprising the following steps: Image acquisition steps: The control mechanism coordinates the illumination mechanism, imaging mechanism, and sample manipulation mechanism to acquire two-dimensional image sequences of the target under different illumination angles and different rotation angles; Three-dimensional morphology reconstruction step: The two-dimensional image sequence is processed to reconstruct the three-dimensional morphology and internal structure information of the target pellet; Defect analysis steps: Based on the reconstruction results, extract the surface contour features of the target pellet, and identify and mark defects by comparing them with the standard radius.

[0012] Furthermore, the image acquisition step includes: controlling the LED array light source to light up in a preset sequence to provide illumination at different angles, and synchronously controlling the rotary motor to step and rotate the target, so that the camera can acquire a low-resolution intensity image under each illumination-rotation angle combination.

[0013] Furthermore, the three-dimensional topography reconstruction step includes: For each rotation angle, the Fourier stacked imaging algorithm is applied to the image sequence to recover the two-dimensional complex amplitude projection of the target at that angle; Using two-dimensional complex amplitude projections at different rotation angles as input, the three-dimensional refractive index distribution of the target pellet is calculated through a tomographic reconstruction algorithm, thereby obtaining three-dimensional morphology and structural information.

[0014] Furthermore, the defect analysis step includes: calculating the Euclidean distance from each point on the surface contour of the three-dimensional shape to the center of the fitted circle as the actual radius, comparing the actual radius with the standard radius, and marking the point as a defect point if the deviation exceeds a preset threshold.

[0015] Furthermore, the preset threshold is 1.11%.

[0016] The present invention provides an integrated device and method for three-dimensional imaging detection of carbon-deuterium targets, which has the following significant advantages compared with the prior art: First, in terms of imaging capabilities, this invention achieves true three-dimensional non-destructive, high-contrast imaging. By fusing Fourier layered imaging with optical tomography algorithms, the three-dimensional refractive index distribution inside the target pellet can be directly reconstructed without the need for refractive index matching fluid. This solves the problem of low contrast in X-ray CT for light element samples and allows for intuitive observation of the three-dimensional structure, providing an information depth unmatched by traditional two-dimensional images.

[0017] Secondly, in terms of detection performance, it achieves high precision and quantification. The device can provide spatial coordinates, size, and shape parameters accurate to the nanometer level, rather than qualitative image judgment, providing a reliable quantitative basis for target quality.

[0018] Finally, breakthroughs have been achieved in system efficiency and ease of use. This invention deeply integrates optics, mechanics, electronics, and control, realizing fully automated one-click detection. Users only need to place the sample and start the program, and the system can automatically complete illumination, acquisition, motion control, 3D reconstruction, and defect analysis, significantly reducing the difficulty of operation and reliance on professional personnel. The detection efficiency is significantly improved compared to traditional methods that rely heavily on manual intervention, providing a stable and efficient measurement platform for scientific research and industrial applications. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the device structure of the present invention; Figure 2 This is a schematic diagram of the software control interface of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0021] This embodiment provides an integrated three-dimensional imaging detection device for carbon-deuterium targets, specifically including an illumination mechanism, an imaging mechanism, a sample manipulation mechanism, and a control mechanism. Figure 1 The overall structure of the device is described below, with specific components and structure as follows.

[0022] The lighting mechanism includes an addressable LED array light source and its driving circuit. The core of the lighting mechanism is the addressable LED array light source. In actual construction, a 15×15 LED matrix is ​​selected, with each LED unit having a center wavelength of 625nm (red light band) and a center-to-center spacing of 4mm between LED units. The LED array driving circuit adopts a constant current driving mode, and by receiving PWM (pulse width modulation) signals from the control mechanism, it can precisely control the on / off state and luminous intensity of each LED unit. This driving circuit is connected to the control mechanism via a ribbon cable to ensure the stability of signal transmission.

[0023] The imaging mechanism is optically connected to the illumination mechanism and includes a microscopic optical system and a camera. The microscopic optical system consists of an objective lens and a simplified mirror arranged in a 4f optical path structure. In this 4f optical path structure, the distance from the sample to the front focal plane of the objective lens is the focal length of the objective lens; the distance from the rear focal plane of the objective lens to the front focal plane of the simplified mirror is the sum of the focal lengths of the objective lens and the simplified mirror; and the distance from the rear focal plane of the simplified mirror to the target surface of the camera is the focal length of the simplified mirror. In specific implementation, the following distances need to be precisely calibrated: The sample stage plane to the microscope objective (preferably OLYMPUS PLN4X, numerical aperture NA=0.10, focal length...) The distance to the front focal plane should be exactly equal to the focal length of the objective lens. .

[0024] The distance from the back focal plane of the microscope objective to the simplified lens (preferred model is THORLABS-TTL180-A, focal length) is... =180mm) The distance between the front focal planes should be exactly equal to the sum of the two focal lengths. + ).

[0025] The distance from the back focal plane of the simplified lens to the target surface of the camera (preferably a Basler ace a2A4504-18umPRO) should be exactly equal to the focal length of the simplified lens. .

[0026] This 4F structure ensures zero coma and astigmatism across the entire imaging field of view, forming the basis for achieving high-resolution (≤980nm) imaging. The camera connects to the control unit via USB 3.0 or Camera Link interface for high-speed transmission of image data.

[0027] The sample handling mechanism is key to achieving automated detection, and its mechanical connection relationship is as follows: The adsorption rotation unit, consisting of a rotary motor and a transparent suction nozzle mechanically connected, is used to fix and rotate the target pellet. Specifically, the output shaft of the rotary motor (such as a coreless motor) is rigidly connected to the transparent suction nozzle (such as glass or high-strength optical plastic) via a precision coupling. During installation, calibration must be performed using tools such as a dial indicator to ensure that the coaxiality error between the central axis of the suction nozzle and the rotation axis of the rotary motor is controlled within ±5μm. The suction nozzle is connected to a miniature vacuum pump via a hose, adsorbing the target pellet through negative pressure.

[0028] The displacement and positioning unit, mechanically connected by a three-dimensional displacement stage and a rotary stage, is used to control the spatial position of the adsorption rotation unit. Specifically, a high-precision three-dimensional electric displacement stage (e.g., driven by a lead screw or piezoelectric ceramic, with a positioning accuracy of 1 μm) serves as the base platform. An electric rotary stage (angular positioning accuracy of 0.01°) is mounted on it. The adsorption rotation unit is fixed to the central turntable of this rotary stage via a mounting bracket. Thus, the displacement stage controls the movement of the adsorption unit in the X, Y, and Z directions, while the rotary stage controls its rotation around the Z-axis.

[0029] The loading unit, with its layered structure and built-in tracks, is used to store target pellets to be tested and those already tested. The loading unit includes layered test chambers and tested chambers, with internal tracks guiding the target pellets between these chambers. In this embodiment, the loading stage body is made of metal or high-strength engineering plastic and is clearly divided into upper and lower layers: the upper layer is the test pellet chamber, and the lower layer is the tested pellet chamber. Each layer has precisely machined V-shaped or U-shaped tracks to constrain the rolling path of the target pellets, ensuring they move along a predetermined track to the adsorption station.

[0030] The control mechanism is electrically connected to the lighting mechanism, imaging mechanism, and sample manipulation mechanism, and includes a main controller and a slave controller communicating with the main controller. The output of the slave controller is connected to the driving circuit of the LED array light source and the rotary motor, respectively. The main controller is also communicatively connected to the displacement stage controller and the camera. Specifically, the control mechanism adopts a master-slave architecture and is responsible for the coordinated operation of the entire system, including: From the controller: the core is a microcontroller, such as the Arduino Mega 2560 microcontroller. Its digital output pins (D2-D13, etc.) are connected to the LED array via a driver circuit board to control the lighting timing. Another set of its digital output pins are connected to the rotary motor and the stepper motor of the displacement stage via a motor driver module (such as the A4988 stepper motor driver).

[0031] Main controller: This is an industrial computer (host computer). It communicates with the Arduino microcontroller via a USB serial cable to send control commands. Simultaneously, it connects to a camera via another USB interface to trigger image acquisition. Furthermore, it communicates with the controllers integrated into the displacement stage and rotary stage via an Ethernet port or RS232 interface to send position and angle commands.

[0032] The output of the slave controller (Arduino) is physically connected to the input interface of the LED driver circuit and the drive interface of the rotary motor via wires. The master controller establishes communication connections with the displacement stage controller, the camera, and the slave controller via data lines. All these connections constitute a complete control network.

[0033] The specific implementation process is as follows. Based on the aforementioned integrated three-dimensional imaging and detection device for carbon-deuterium targets, the device's software control interface ( Figure 2 To perform the operation: 1. Initialization and parameter settings The user first enters the detection parameters in the "Parameter Settings" area of ​​the software interface. These parameters include: Optical parameters: illumination source wavelength (λ=625nm), physical size of LED array and spacing between LEDs, camera pixel size (2.74μm), objective lens magnification (4X) and numerical aperture (0.10).

[0034] Motion parameters: total angle range of target rotation (0-180°) and step angle (e.g., 5° or 10°).

[0035] Acquisition parameters: camera exposure time (usually 50-100ms), gain, and the size of the image to be acquired.

[0036] 2. Automated data collection process After the user clicks the "Start Detection" button, the software executes the following collaborative control according to the preset procedure: The main controller sends commands to the displacement stage and the rotary stage to move the target pellet from the test chamber of the stage and precisely position it to the focal plane of the microscopic system.

[0037] For each rotation angle (where i is the angle index) (representing the i-th rotation angle), the main controller commands the Arduino to light up each LED unit in the LED array according to a predetermined sequence (such as line-by-line scanning).

[0038] Each time an LED is lit, the main controller triggers the camera to capture a corresponding low-resolution intensity image. Where (x,y) are the image pixel coordinates. This indicates that the LED is illuminated in the m-th row and n-th column, with a rotation angle of . The light intensity value collected at that time.

[0039] that angle After all the LED images have been acquired, the main controller commands the rotary table to step to the next angle. Repeat the above process until all set angles are covered.

[0040] During this process, users can observe the real-time images captured by the camera in the "Real-time Imaging" section of the interface, and monitor the system status through the "Memory Usage" and "Reconstruction Progress" display bars.

[0041] This embodiment provides a three-dimensional imaging detection method for carbon-deuterium targets. The method employs an integrated three-dimensional imaging detection device for carbon-deuterium targets, comprising three sequentially executed steps: image acquisition, three-dimensional morphology reconstruction, and defect analysis. The specific process is described below.

[0042] 1. Image Acquisition Step: The control mechanism coordinates the illumination mechanism, imaging mechanism, and sample manipulation mechanism to acquire a two-dimensional image sequence of the target pellet under different illumination angles and different rotation angles. Specifically, the image acquisition step includes: controlling the LED array light source to light up in a preset sequence to provide illumination at different angles, and simultaneously controlling the rotary motor to step and rotate the target pellet, with the camera acquiring a low-resolution intensity image under each illumination-rotation angle combination.

[0043] 2. Three-dimensional morphology reconstruction step: The two-dimensional image sequence is processed to reconstruct the three-dimensional morphology and internal structure information of the target pellet. The three-dimensional morphology reconstruction step includes: applying a Fourier layered imaging algorithm to the image sequence acquired at each rotation angle to recover the two-dimensional complex amplitude projection of the target pellet at that angle; using the two-dimensional complex amplitude projection at different rotation angles as input, and calculating the three-dimensional refractive index distribution of the target pellet through a tomographic reconstruction algorithm to obtain the three-dimensional morphology and structural information.

[0044] (1) Complex amplitude projection restoration (Fourier stacked imaging algorithm) For a series of low-resolution images acquired at each rotation angle θ The high-resolution complex amplitude projection is recovered using the following iterative process: Initialization: Select an image acquired perpendicular to the illumination (i.e., when the center LED bead of the LED array is lit), upsample it (e.g., bilinear interpolation to 1000×1000 pixels), then perform a Fourier transform, and multiply it with the pupil function of the objective lens (a circular low-pass filter) to obtain an initial estimate of the object's spectrum. (where (u, v) are frequency domain coordinates, (This represents the object spectrum estimated in the 0th iteration).

[0045] Iterative update (taking the k-th iteration as an example): a. For the (m, n)th LED, calculate the frequency domain offset corresponding to its illumination wave vector. ( and These are the offsets of the spectrum on the u-axis and v-axis, respectively.

[0046] b. Estimation from the current object's spectrum Extract the corresponding sub-region spectrum from (representing the object spectrum after the k-th iteration update).

[0047] c. Perform an inverse Fourier transform on the spectrum of this sub-region to obtain the spatial complex amplitude distribution. (where (x,y) are spatial coordinates;) This represents the estimate of the complex amplitude distribution of the object obtained under LED illumination in the m-th row and n-th column during the k-th iteration.

[0048] d. Apply spatial constraints: maintain The phase remains unchanged, and its amplitude is replaced with the square root of the amplitude of the actual acquired image, thus completing one spatial update.

[0049] e. Perform another Fourier transform on the updated spatial distribution and update it back to the object spectrum. The corresponding sub-region.

[0050] Iterate through all LED lighting angles to complete one global iteration. Typically, after 10-20 iterations, the spectrum estimation will converge, at which point the final object spectrum will be obtained. By performing an inverse Fourier transform, a high-resolution complex amplitude projection at the rotation angle θ can be obtained. ,(in, This represents the amplitude projection distribution recovered under a rotation angle θ. (This represents the phase projection distribution recovered under a rotation angle θ).

[0051] (2) Reconstruction of three-dimensional refractive index distribution (tomography algorithm) Projection data conversion: Recovering the phase projection at each rotation angle θ This is converted into optical projection data for tomographic inversion. According to the theory of light propagation in weakly absorbing media, the phase projection is linearly related to the fluctuations of the real part of the sample's refractive index. Specifically, it is converted into projection data related to the real part of the refractive index. The calculation formula is as follows: Where k is the wave number, and its value is... λ is the center wavelength of the illumination (i.e., the center wavelength of the LED light source, for example, 625nm). It can be viewed as the projection of the offset of the real part of the refractive index relative to the background refractive index at angle θ.

[0052] Tomographic inversion: combining different perspectives A series of projection data obtained below As a sinogram (i.e., projection dataset), a filtered back projection (FBP) algorithm is used for 3D reconstruction: first, a ramp filter (such as a Ram-Lak filter) is applied to the projection data at each angle for frequency domain filtering to eliminate the blurring caused by simple back projection; then, the filtered projection data is back-projected back into the 3D spatial grid along its corresponding projection direction; finally, by accumulating the back projection data from all angles, the 3D refractive index offset distribution inside the sample is reconstructed. .

[0053] Finally, according to the relation The three-dimensional refractive index distribution of the sample was calculated. .in, The refractive index is the refractive index of the medium surrounding the sample (such as air), usually taken as... Therefore, it can be simplified to .

[0054] 3. Defect Analysis Steps: Based on the reconstruction results, the surface contour features of the target pellet are extracted, and defects are identified and marked by comparing them with the standard radius. The defect analysis steps include: calculating the Euclidean distance from each point on the surface contour of the three-dimensional morphology to the center of the fitted circle as the actual radius; comparing the actual radius with the standard radius; and marking the point as a defect point if the deviation exceeds a preset threshold.

[0055] (1) Surface contour extraction The reconstructed three-dimensional refractive index distribution The process is performed layer by layer (slice by slice) along the Z-axis. For each 2D slice, its gradient magnitude is calculated. Threshold segmentation or edge detection algorithms (such as the Canny algorithm) are used to extract the internal and external contours of the target pellet in this layer, thereby obtaining the point cloud data of the entire target pellet's internal and external surfaces.

[0056] (2) Defect identification and judgment For the obtained point clouds of the inner and outer surfaces, the least squares method is used to fit an ideal circle center coordinate. This point is the geometric center of the target in three-dimensional space. Calculate the value of each point on the surface. The Euclidean distance to the center of the fitted circle is used as the actual radius of that point. (That is, the distance between this surface point and the geometric center). With the standard design radius of the target pellet (Compare with the nominal radius value determined according to the product specifications of the target pellet). Set a relative tolerance threshold δ to define the allowable radius deviation range. For any surface point, if its radius deviation meets the following... If the relative error between the actual radius and the standard radius exceeds a threshold, the point is marked as a defect. In a preferred embodiment of the invention, this threshold δ is set to 1.11%. Finally, the software highlights all defect points on the 3D model and generates an inspection report containing the number, location, and size of defects, automatically determining whether the sample is qualified.

[0057] In summary, this invention enables automatic, quantitative, and high-resolution three-dimensional imaging and evaluation of the morphology and internal defects of target pellets.

[0058] Embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of one or more computer-usable storage media (including, but not limited to, disk storage, etc.) containing computer-usable program code. CD - ROM It takes the form of a computer program product implemented on (such as optical memory, etc.).

[0059] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0060] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0061] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0062] Contents not described in detail in this specification are prior art known to those skilled in the art. It is hereby indicated that the above description is intended to help those skilled in the art understand this invention, but does not limit the scope of protection of this invention. Any equivalent substitutions, modifications, improvements, or simplifications of the above descriptions that do not depart from the essential content of this invention fall within the scope of protection of this invention.

Claims

1. A three-dimensional imaging and detection integrated device for carbon-deuterium targets, characterized in that, include: Lighting mechanism, including addressable LED array light source and its driving circuitry; An imaging mechanism, optically connected to the illumination mechanism, includes a microscopic optical system and a camera, wherein the microscopic optical system consists of objective lenses and telescopes arranged in a 4f optical path structure; The sample handling mechanism includes: The adsorption rotation unit, consisting of a rotary motor and a transparent suction nozzle mechanically connected, is used to fix and rotate the target pellet. The displacement positioning unit, consisting of a three-dimensional displacement stage and a rotary stage mechanically connected, is used to control the spatial position of the adsorption rotation unit. The cargo unit is layered and has built-in tracks for storing targets to be tested and those already tested. The control mechanism is electrically connected to the lighting mechanism, imaging mechanism, and sample manipulation mechanism, and includes a main controller and a slave controller that communicates with the main controller. The output of the slave controller is connected to the driving circuit of the LED array light source and the rotary motor, respectively. The main controller is also communicatively connected to the displacement stage controller and the camera.

2. The integrated three-dimensional imaging and detection device for carbon-deuterium targets according to claim 1, characterized in that, In the adsorption rotation unit, the central axis of the transparent suction nozzle is coaxial with the rotation axis of the rotary motor.

3. The integrated three-dimensional imaging and detection device for carbon-deuterium targets according to claim 1, characterized in that, The carrying unit includes a layered chamber for testing and a chamber that has already been tested, and an internal track is provided to guide the target pellet to move between the chamber for testing and the chamber that has already been tested.

4. The integrated three-dimensional imaging and detection device for carbon-deuterium targets according to claim 1, characterized in that, In the 4f optical path structure, the distance from the sample to the front focal plane of the objective lens is the focal length of the objective lens, the distance from the rear focal plane of the objective lens to the front focal plane of the simplified lens is the sum of the focal lengths of the objective lens and the simplified lens, and the distance from the rear focal plane of the simplified lens to the target surface of the camera is the focal length of the simplified lens.

5. The integrated three-dimensional imaging and detection device for carbon-deuterium targets according to claim 1, characterized in that, The slave controller includes a microcontroller, which is an Arduino microcontroller.

6. A three-dimensional imaging detection method for carbon-deuterium targets, characterized in that, The method employs the integrated three-dimensional imaging and detection device for carbon-deuterium targets as described in any one of claims 1-5, and includes the following steps: Image acquisition steps: The control mechanism coordinates the illumination mechanism, imaging mechanism, and sample manipulation mechanism to acquire two-dimensional image sequences of the target under different illumination angles and different rotation angles; Three-dimensional morphology reconstruction step: The two-dimensional image sequence is processed to reconstruct the three-dimensional morphology and internal structure information of the target pellet; Defect analysis steps: Based on the reconstruction results, extract the surface contour features of the target pellet, and identify and mark defects by comparing them with the standard radius.

7. The three-dimensional imaging detection method for carbon-deuterium targets according to claim 6, characterized in that, The image acquisition step includes: controlling the LED array light source to light up in a preset sequence to provide illumination at different angles, and simultaneously controlling the rotary motor to step and rotate the target, with the camera acquiring a low-resolution intensity image under each illumination-rotation angle combination.

8. The three-dimensional imaging detection method for carbon-deuterium targets according to claim 6, characterized in that, The three-dimensional topography reconstruction steps include: For each rotation angle, the Fourier stacked imaging algorithm is applied to the image sequence to recover the two-dimensional complex amplitude projection of the target at that angle; Using two-dimensional complex amplitude projections at different rotation angles as input, the three-dimensional refractive index distribution of the target pellet is calculated through a tomographic reconstruction algorithm, thereby obtaining three-dimensional morphology and structural information.

9. A three-dimensional imaging detection method for carbon-deuterium targets according to claim 8, characterized in that, The defect analysis step includes: calculating the Euclidean distance from each point on the surface contour of the three-dimensional shape to the center of the fitted circle as the actual radius, comparing the actual radius with the standard radius, and marking the point as a defect point if the deviation exceeds a preset threshold.

10. A three-dimensional imaging detection method for carbon-deuterium targets according to claim 9, characterized in that, The preset threshold is 1.11%.

Images