A method of bright field imaging based on a dmd

Abstract

This invention discloses a strong backlight imaging method based on a DMD (Digital Modulator-Demiller), belonging to the field of optical imaging technology. This method involves spatial pixel-level matching and synchronous triggering feedback between the camera and the DMD. Under strong backlight imaging conditions, the strong backlight passing through the target and the light reflected from the target together form the incident light, which sequentially passes through the imaging lens, the DMD target surface, the relay matching imaging lens, and the camera target surface. The overall grayscale value is reduced by the DMD mask to obtain an initial image containing target contour information. Based on this image, an initial mask is generated for differential grayscale modulation of the target and background areas. Further, local overexposed areas are detected and pixel-level masks are generated. Simultaneously, a high bit-depth grayscale modulation mode of the DMD is introduced to finely suppress brightness in locally bright areas, ultimately obtaining an image of the target without overexposure and with clear details. This invention can effectively suppress overexposure and improve imaging contrast under strong backlight conditions, and is suitable for high dynamic range backlight imaging scenarios.

Description

Technical Field

[0001] This invention belongs to the field of optical imaging technology, specifically relating to a strong backlight imaging method based on DMD. Background Technology

[0002] In space, the light energy of stars and other celestial bodies is extremely powerful, posing a significant challenge to imaging systems when imaging celestial bodies or particles. Especially in backlit conditions, traditional imaging methods struggle to effectively suppress the intense light from the source, leading to overexposure and consequently affecting image quality. Due to the vast brightness difference between the light source and the target, many stars, particles, or other celestial objects can only be captured with complete outline information under backlit conditions.

[0003] Traditional imaging techniques, such as filters, apertures, or fixed blocking devices, can suppress some strong light, but their effectiveness is limited and they cannot flexibly cope with different light intensity variations. At the same time, existing imaging systems often rely on a single optical element, making it difficult to precisely control the direction and intensity of light transmission. This makes it difficult to simultaneously ensure high contrast and high dynamic range when imaging in strong backlight environments.

[0004] In addition, current astronomical telescopes or space exploration equipment generally face the difficulty of observing celestial objects against the background of strong starlight or sunlight. Such strong light illumination results in low image contrast of the imaging system, making it impossible to accurately capture detailed information, and there are even cases where some targets or particles are "submerged" due to strong light.

[0005] A Digital Micromirror Device (DMD) is a spatial light modulator based on microelectromechanical systems (MEMS). It consists of an array of numerous independently controllable micromirror units, each of which can switch between different tilt states to directionally reflect and modulate incident light. By controlling the on / off state or grayscale state of each micromirror unit, precise spatial modulation of light intensity can be achieved. DMDs not only possess high spatial resolution and modulation accuracy but also offer advantages such as fast response speed and programmable control. They can achieve multi-level grayscale modulation or equivalent grayscale modulation, thus meeting the requirements for dynamic range adjustment of light intensity under strong backlight conditions.

[0006] Therefore, to meet the demand for high-quality imaging in strong backlight environments, there is an urgent need for a technical solution that can precisely adjust light intensity and improve image quality under backlight conditions. This invention introduces a digital micromirror device (DMD) and spatial grayscale modulation technology, combined with an adaptive algorithm, to achieve effective control of light intensity. This enables the successful acquisition of clear target outlines and improved dynamic range of imaging under strong backlight conditions, thus overcoming the shortcomings of traditional imaging techniques. Summary of the Invention

[0007] To address the problems of overexposure and loss of image details in existing strong backlight imaging methods, this invention provides a strong backlight imaging method based on DMD. By introducing pixel-level matching grayscale adjustment and feedback correction mechanisms into the imaging system, high-contrast imaging under backlight conditions is achieved.

[0008] The technical solution of the present invention to solve the above-mentioned technical problems is:

[0009] A DMD-based method for imaging strong backlighting includes the following steps:

[0010] S1. Construct a strong backlight imaging system. This imaging system includes a digital micromirror device (DMD), a camera, a processor, an imaging lens, and a relay matching imaging lens. The DMD, camera, and processor are connected in pairs. The light entering the system passes through the imaging lens, the DMD target surface, the relay matching imaging lens, and the camera target surface in sequence.

[0011] S2. Perform spatial calibration on the camera and DMD to obtain the mapping matrix H between the camera imaging target surface and the DMD target surface;

[0012] S3. Under strong backlight imaging conditions, the strong backlight passing through the target and the reflected light from the target surface together form incident light. After passing through the strong backlight imaging system, the camera captures an overexposed image image_1.

[0013] S4. Adjust the overall grayscale value of the DMD to modulate the incident light, thereby reducing the intensity of the reflected light collected by the camera, and the camera captures an initial image image_2 containing the target contour information.

[0014] S5. Process image_2 using algorithms and extract the mask to obtain the target mask mask_1. Then, use the pixel matching magnification M of the relay matching imaging lens and the mapping relationship matrix H to scale and correct the target mask to the DMD target surface size, generating the initial mask image mask_2, which satisfies the following relationship:

[0015] ;

[0016] S6. The grayscale value of the differential mask image mask_2 is loaded into the DMD for modulation, and the camera captures an image image_3 with a bright background and uneven target brightness.

[0017] S7. Set the brightness threshold to gray_max, detect overexposed areas in image_3 whose gray values ​​exceed gray_max, reduce the gray values ​​at the corresponding positions in mask image_2 and perform DMD modulation, camera shooting, and processor processing. Iterate through the process of DMD modulation, camera image acquisition, and mask update. The termination condition of the iteration process is that the bright areas in the image meet the preset brightness range, generate pixel-level mask image_3, and capture image_4 with clear details and no overexposure.

[0018] Preferably, in step S1, the imaging lens images the target onto the DMD target surface, and the relay matching imaging lens is used to achieve pixel magnification matching between the DMD target surface and the camera target surface; the digital micromirror device (DMD), the camera, and the processor are connected in pairs, so that the camera synchronously captures images after receiving the trigger signal from the DMD, the processor acquires the images and performs algorithm processing on them, creates a mask, and loads it onto the DMD; the DMD target surface and the camera target surface are set parallel to each other, the incident light makes an angle θ with the normal of the DMD target surface, satisfying θ=2φ, where φ is the deflection angle of the DMD micromirror, and the reflected light is perpendicular to the DMD target surface and the camera target surface.

[0019] Preferably, in step S1, the digital micromirror device (DMD) includes an array structure composed of multiple independently controllable micromirror units, which modulate and reflect the incident light by deflection; the camera is a CMOS camera or a CCD camera; the processor is at least one of a central processing unit (CPU), graphics processing unit (GPU), microprocessor unit (MPU), digital signal processor (DSP), microcontroller (MCU), system-on-a-chip (SoC), application-specific integrated circuit (ASIC), image signal processor (ISP), neural network processor (NPU), field-programmable gate array (FPGA), or complex programmable logic device (CPLD); the imaging lens and the relay matching imaging lens are optical imaging devices composed of a lens group with imaging function and a mechanical housing.

[0020] Preferably, in step S2, the mapping matrix H is used to establish the correspondence between the camera imaging coordinates and the DMD target surface coordinates, including affine transformation, perspective transformation, affine transformation with distortion correction, or spatial coordinate mapping.

[0021] Preferably, in step S3, the combination of strong backlight and reflected light to form incident light includes coaxial superposition or non-coaxial superposition.

[0022] Preferably, in step S4, the overall grayscale value is a preset or dynamically adjusted grayscale parameter used to adjust the overall intensity distribution of the incident light, enabling the camera to acquire an image containing target contour information.

[0023] Preferably, in step S5, the target mask is obtained by image processing methods based on threshold segmentation, edge detection, or region segmentation; the pixel matching ratio M can be adjusted according to the relationship between the DMD pixel size and the camera pixel size.

[0024] Preferably, in step S6, the gray values ​​in the differential mask image mask_2 are processed in different regions, including the target region, the background region, or multiple regions divided based on image features, so as to achieve differential gray-scale modulation of the target region and the background region, thereby improving the imaging contrast of the target under strong backlight conditions.

[0025] Preferably, in step S7, the pixel-level mask image mask_3 adopts a multi-level grayscale modulation method to perform fine brightness suppression on local bright areas. The multi-level grayscale modulation includes grayscale representation methods with different grayscale bit depths, equivalent grayscale control methods achieved through multi-frame modulation, or continuous grayscale modulation methods based on spatial distribution.

[0026] Compared with the prior art, the present invention has the following advantages:

[0027] 1. By constructing a strong backlight imaging system including a digital micromirror device (DMD), a camera, and a relay matching imaging lens, and realizing the pixel-level matching relationship between the DMD target surface and the camera target surface, spatial modulation information can be accurately applied to the corresponding imaging pixels, thereby significantly improving the spatial accuracy and imaging consistency of light intensity modulation.

[0028] 2. This invention utilizes DMD to perform overall grayscale modulation of incident light, effectively reducing the intensity of incident light under strong backlight conditions, enabling the imaging system to acquire an initial image containing target contour information, thus solving the problems of overexposure and difficulty in extracting target contours in traditional imaging methods under strong light conditions.

[0029] 3. This invention extracts a mask from the initial image and performs spatial mapping and correction on the mask using a mapping relationship matrix, thereby achieving the generation of a spatial modulation mask consistent with the DMD target surface, thus ensuring an accurate correspondence between the modulation area and the actual imaging area.

[0030] 4. This invention achieves independent control of the target area and the background area by setting different gray levels for different regions in the modulation mask. It can suppress overexposure of the target area while ensuring the brightness of the background, thereby effectively improving the image contrast and detail performance.

[0031] 5. This invention achieves a closed-loop modulation process based on image feedback by detecting the brightness of the imaging results and iteratively updating the modulation mask corresponding to the local bright areas. This can gradually suppress local overexposure and improve the brightness uniformity of the imaging results.

[0032] 6. This invention employs pixel-level mask modulation and multi-level grayscale modulation to perform fine brightness control on local bright areas, further enhancing the ability to suppress strong light areas and thus improving the dynamic range of the system.

[0033] 7. This invention achieves adaptive optimization of imaging results through multiple cycles of modulation and imaging processes, with a preset brightness range as the termination condition. This results in images that are both free from overexposure and have clear detail, thus improving imaging quality and stability under strong backlight conditions.

[0034] 8. The imaging system and method proposed in this invention are particularly suitable for high dynamic range backlight imaging scenarios such as stellar-level strong light backgrounds and space particle observation, and have high engineering application value. Attached Figure Description

[0035] Figure 1 This is a flowchart of the strong backlight imaging method based on DMD of the present invention.

[0036] Figure 2 This is a system optical path diagram of the DMD-based strong backlight imaging method of the present invention.

[0037] Figure 3 This is a diagram showing the connection and calibration of the camera and DMD in the DMD-based strong backlight imaging method of the present invention. Detailed Implementation

[0038] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0039] See Figure 1 The DMD-based strong backlight imaging method of the present invention includes the following steps:

[0040] S1. Construct a strong backlight imaging system. This imaging system includes a digital micromirror device (DMD), a camera, a processor, an imaging lens, and a relay matching imaging lens. The DMD, camera, and processor are connected in pairs. The light entering the system passes through the imaging lens, the DMD target surface, the relay matching imaging lens, and the camera target surface in sequence.

[0041] S2. Perform spatial calibration on the camera and DMD to obtain the mapping matrix H between the camera imaging target surface and the DMD target surface;

[0042] S3. Under strong backlight imaging conditions, the strong backlight passing through the target and the reflected light from the target surface together form incident light. After passing through the strong backlight imaging system, the camera captures an overexposed image image_1.

[0043] S4. Adjust the overall grayscale value of the DMD to modulate the incident light, thereby reducing the intensity of the reflected light collected by the camera, and the camera captures an initial image image_2 containing the target contour information.

[0044] S5. Process image_2 using algorithms and extract the mask to obtain the target mask mask_1. Then, use the pixel matching magnification M of the relay matching imaging lens and the mapping relationship matrix H to scale and correct the target mask to the DMD target surface size, generating the initial mask image mask_2, which satisfies the following relationship:

[0045] ;

[0046] S6. The grayscale value of the differential mask image mask_2 is loaded into the DMD for modulation, and the camera captures an image image_3 with a bright background and uneven target brightness.

[0047] S7. Set the brightness threshold to gray_max, detect overexposed areas in image_3 whose gray values ​​exceed gray_max, reduce the gray values ​​at the corresponding positions in mask image_2 and perform DMD modulation, camera shooting, and processor processing. Iterate through the process of DMD modulation, camera image acquisition, and mask update. The termination condition of the iteration process is that the bright areas in the image meet the preset brightness range, generate pixel-level mask image_3, and capture image_4 with clear details and no overexposure.

[0048] See Figures 2-3 The imaging lens images the target onto the DMD target surface. The relay matching imaging lens is used to achieve pixel magnification matching between the DMD target surface and the camera target surface. The pixel matching magnification M can be adjusted according to the relationship between the DMD pixel size and the camera pixel size. The digital micromirror device (DMD), the camera, and the processor are connected in pairs, so that the camera can synchronously capture images after receiving the trigger signal from the DMD. The processor acquires the images and performs algorithm processing on them, creates a mask, and loads it onto the DMD. The DMD target surface and the camera target surface are set parallel to each other. The DMD micromirror deflection angle is 12°, the angle between the incident light and the normal of the DMD target surface is 24°, and the reflected light is perpendicular to the DMD target surface and the camera target surface.

[0049] See Figure 2 The combination of strong backlight and reflected light to form incident light can be coaxial or non-coaxial, and non-coaxial superposition is used in this embodiment.

[0050] See Figure 3The mapping relationship is used to establish the correspondence between the camera imaging coordinates and the DMD target surface coordinates. The mapping relationship includes, but is not limited to, affine transformation, perspective transformation, affine with distortion correction or spatial coordinate mapping. In this embodiment, affine transformation is used.

[0051] The above are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above content. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A strong backlight imaging method based on DMD, characterized in that, Includes the following steps: S1. Construct a strong backlight imaging system. This imaging system includes a digital micromirror device (DMD), a camera, a processor, an imaging lens, and a relay matching imaging lens. The DMD, camera, and processor are connected in pairs. The light entering the system passes through the imaging lens, the DMD target surface, the relay matching imaging lens, and the camera target surface in sequence. S2. Perform spatial calibration on the camera and DMD to obtain the mapping matrix H between the camera imaging target surface and the DMD target surface; S3. Under strong backlight imaging conditions, the strong backlight passing through the target and the reflected light from the target surface together form incident light. After passing through the strong backlight imaging system, the camera captures an overexposed image image_1. S4. Adjust the overall grayscale value of the DMD to modulate the incident light, thereby reducing the intensity of the reflected light collected by the camera, and the camera captures an initial image image_2 containing the target contour information. S5. Process image_2 using algorithms and extract the mask to obtain the target mask mask_1. Then, use the pixel matching magnification M of the relay matching imaging lens and the mapping relationship matrix H to scale and correct the target mask to the DMD target surface size, generating the initial mask image mask_2, which satisfies the following relationship: ； S6. The grayscale value of the differential mask image mask_2 is loaded into the DMD for modulation, and the camera captures an image image_3 with a bright background and uneven target brightness. S7. Set the brightness threshold to gray_max, detect overexposed areas in image_3 whose gray values ​​exceed gray_max, reduce the gray values ​​at the corresponding positions in mask image_2 and perform DMD modulation, camera shooting, and processor processing. Iterate through the process of DMD modulation, camera image acquisition, and mask update. The termination condition of the iteration process is that the bright areas in the image meet the preset brightness range, generate pixel-level mask image_3, and capture image_4 with clear details and no overexposure.

2. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S1, the imaging lens images the target onto the DMD target surface, and the relay matching imaging lens is used to achieve pixel magnification matching between the DMD target surface and the camera target surface; the digital micromirror device (DMD), the camera, and the processor are connected in pairs, so that the camera can synchronously capture images after receiving the trigger signal from the DMD, the processor acquires the images and performs algorithm processing on them, creates a mask, and loads it onto the DMD; the DMD target surface and the camera target surface are set parallel to each other, the incident light makes an angle θ with the normal of the DMD target surface, which satisfies θ=2φ, where φ is the deflection angle of the DMD micromirror, and the reflected light is perpendicular to the DMD target surface and the camera target surface.

3. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S1, the digital micromirror device (DMD) includes an array structure composed of multiple independently controllable micromirror units. The micromirror units modulate the grayscale of the incident light and reflect it by deflection. The camera is a CMOS camera or a CCD camera. The processor is at least one of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor unit (MPU), a digital signal processor (DSP), a microcontroller (MCU), a system-on-a-chip (SoC), an application-specific integrated circuit (ASIC), an image signal processor (ISP), a neural network processor (NPU), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD). The imaging lens and the relay matching imaging lens are optical imaging devices composed of a lens group with imaging function and a mechanical housing.

4. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S2, the mapping matrix H is used to establish the correspondence between the camera imaging coordinates and the DMD target surface coordinates, including affine transformation, perspective transformation, affine transformation with distortion correction, or spatial coordinate mapping.

5. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S3, the combination of strong backlight and reflected light to form incident light can be achieved through coaxial superposition or non-coaxial superposition.

6. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S4, the overall grayscale value is a preset or dynamically adjusted grayscale parameter used to adjust the overall intensity distribution of the incident light, enabling the camera to acquire an image containing target contour information.

7. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S5, the target mask is obtained by image processing methods based on threshold segmentation, edge detection, or region segmentation; the pixel matching ratio M can be adjusted according to the relationship between the DMD pixel size and the camera pixel size.

8. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S6, the gray values ​​in the differential mask image mask_2 are processed in different regions, including the target region, the background region, or multiple regions divided based on image features, in order to achieve differential gray-scale modulation between the target region and the background region, thereby improving the imaging contrast of the target under strong backlight conditions.

9. The DMD-based strong backlight imaging method according to claim 1, characterized in that, In step S7, the pixel-level mask image mask_3 adopts a multi-level grayscale modulation method to perform fine brightness suppression on local bright areas. The multi-level grayscale modulation includes grayscale representation methods with different grayscale bit depths, equivalent grayscale control methods achieved through multi-frame modulation, or continuous grayscale modulation methods based on spatial distribution.

Images