Monochrome sensor hybrid coded illumination and image reconstruction method and imaging system

Abstract

The application discloses a monochrome sensor mixed coding illumination and image reconstruction method and an imaging system, relates to the technical field of computational imaging, and comprises the following steps: first, at least three RGB mixed illumination modes are designed, and a coding matrix is constructed; then, a plurality of frames of gray scale images under the same target state are collected through a global shutter monochrome sensor according to sequence exposure; subsequently, a generalized linear model is constructed based on the coding matrix, an equation set is solved, and a color image is reconstructed; finally, a standard format image is output after post-processing. The application can eliminate dynamic color smearing in principle, has frame-level self-calibration capability, is compatible in algorithm and strong in robustness, keeps the system small and low in cost, and is suitable for dynamic imaging high requirement scenes such as medical and industrial scenes.

Description

Technical Field

[0001] This invention relates to the field of computational imaging technology, and more specifically to a method and imaging system for hybrid coded illumination and image reconstruction using a monochrome sensor. Background Technology

[0002] Existing endoscopic color imaging technology is mainly divided into two core technical routes: First, direct imaging using a color image sensor. This route has inherent defects such as insufficient sensitivity in low light environments and spatial resolution loss due to the design of the color filter array. Second, imaging using a monochrome sensor combined with time-division monochrome illumination. Although this route can improve low light sensitivity and spatial resolution to a certain extent, it has the key technical bottleneck of color motion artifacts in dynamic scenes.

[0003] Specifically, in the time-division monochromatic illumination scheme, light of different spectra needs to illuminate the target scene separately and acquire corresponding grayscale images. Since there is a millisecond-level time difference in the acquisition of images from each channel, when the target scene (such as pulsating organs in the body) or the endoscope probe moves, different color components of the same target will be recorded at different times, resulting in color misalignment and trailing in the reconstructed color image, which seriously affects the accuracy and reliability of scenarios such as surgical navigation and real-time detection of industrial defects.

[0004] Existing improvement technologies attempt to alleviate the above problems by improving lighting switching speed and increasing sensor frame rate through hardware optimization. However, due to physical constraints such as light source response time, sensor readout rate, and data transmission bandwidth, they cannot fundamentally eliminate dynamic artifacts. In particular, the color ghosting problem remains prominent in high-speed motion scenes (such as heartbeat, rapid movement of instruments, and rapid exploration inside pipes).

[0005] In addition, traditional time-sharing lighting schemes rely on a one-time pre-calibration, and the system parameters are fixed after calibration. This makes it impossible to compensate for color distortion caused by factors such as equipment temperature drift, light source aging, and nonlinear shift in sensor response during long-term imaging. As a result, the consistency of imaging color decreases significantly after long-term operation or environmental changes, which further affects the reliability of the application.

[0006] Therefore, there is an urgent need for an endoscopic imaging solution that can fundamentally solve the problem of color motion blur in high-speed motion scenes, has real-time self-calibration capabilities, and maintains system miniaturization, low cost, and high efficiency to fill the gap in existing technologies. Summary of the Invention

[0007] In view of this, the present invention provides a method and imaging system for hybrid coded illumination and image reconstruction using a monochrome sensor, which solves the problems existing in the prior art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: A method for hybrid coded illumination and image reconstruction using a monochrome sensor includes the following steps: S1. Design at least three different RGB mixed lighting modes, in which multiple light sources of different spectra are simultaneously illuminated and the intensity ratio of each spectrum is different, and construct the lighting encoding-color response matrix of the mixed lighting mode; S2. The control lighting system performs at least three consecutive exposures according to a preset mixed lighting mode sequence. During each exposure, a complete grayscale image is acquired by the global shutter monochrome sensor in the corresponding mixed lighting mode, and all exposures are completed under the same target state. S3. Construct a generalized linear model based on the illumination coding-color response matrix, substitute the acquired multi-frame grayscale images into the generalized linear model to construct a system of linear equations, and reconstruct the color image of the target scene by solving the system of linear equations, and calculate the true intensity value of the RGB three channels of each pixel. S4. Post-process the reconstructed color image and output a standard format color image.

[0009] Optionally, in S1, the light source in the RGB mixed lighting mode is a red, green, and blue LED light source, and the number of spectral light sources can be expanded.

[0010] Optionally, in S1, an illumination coding-color response matrix under mixed lighting modes is constructed. Specifically, using a standard color chart as the target, images are acquired under each RGB mixed lighting mode. The spectral response coefficients of the sensor are determined through an optimization algorithm. An illumination coding-color response matrix that characterizes the mapping relationship between the sensor's grayscale response and the true color under different mixed lighting modes is established and stored.

[0011] Optionally, in S3, the optimization algorithm used to solve the linear equation system can be any one of the following: least squares method, weighted least squares method, ridge regression, principal component regression, or singular value decomposition method, which can be flexibly selected according to noise conditions and exposure frame number.

[0012] Optionally, after reconstructing the color image in S3, the color image can be locally optimized using optical flow.

[0013] Optionally, post-processing operations in S4 include one or more combinations of noise reduction, contrast adjustment, edge enhancement, gamma correction, and color space conversion.

[0014] An endoscopic imaging system, employing the monochrome sensor hybrid coded illumination and image reconstruction method described above, includes: a hybrid coded illumination module, a monochrome image sensor module, a control and processing module, an image output module, and an endoscope probe, with each module integrated and packaged inside the endoscope probe; The hybrid coded lighting module includes multiple light sources with different spectral characteristics, an independent and controllable constant current drive circuit, and a light homogenizing component. The constant current drive circuit synchronously drives each light source and independently adjusts the luminous intensity of each spectral light source. The monochrome image sensor module uses a global shutter monochrome CMOS sensor to acquire grayscale images under mixed illumination modes; The control and processing module is based on a programmable logic device or an embedded processor and includes a lighting control unit, an image acquisition control unit, an image reconstruction and processing unit, and a self-calibration unit. The image output module is used to output the reconstructed color image to a display device or storage device.

[0015] Optionally, in the control and processing module, the lighting control unit controls the switching of the mixed lighting mode according to a preset sequence or adaptive strategy, the image acquisition control unit synchronously controls the exposure timing and data reading of the sensor, the image reconstruction processing unit implements a variety of reconstruction algorithms under the framework of a generalized linear model, and the self-calibration unit implements a frame-level self-calibration mechanism and calibrates the system parameters in real time.

[0016] Optionally, the endoscope probe is a compact, integrated imaging unit with a probe diameter ranging from 1mm to 8mm.

[0017] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method and imaging system for hybrid coded illumination and image reconstruction of a monochrome sensor, which has the following beneficial effects: (1) Solving the problem of motion color trailing during time-division illumination of monochrome sensors: In high-speed motion scenes, the present invention can completely avoid the trailing artifacts of color separation, and there may only be slight motion blur determined by the exposure time, providing stable and color-uniform dynamic images. This effect cannot be achieved by traditional time-division illumination technology, and significantly improves the accuracy of dynamic imaging.

[0018] (2) Frame-level self-calibration with excellent color stability: Through the frame-level self-calibration mechanism, the present invention can continuously calibrate the illumination intensity coefficient and sensor response coefficient during system operation, effectively compensate for color distortion caused by factors such as equipment temperature drift, light source aging, and sensor response nonlinear offset, ensure the consistency of imaging color at different times and under different environments, and improve the long-term reliability of the system.

[0019] (3) The algorithm framework is open and robust: the generalized linear model framework can be compatible with a variety of robust estimation algorithms, and can be flexibly selected according to the requirements of computing power, frame rate and imaging accuracy, adapting to different noise conditions and application scenarios, while expanding the scope of patent protection and having strong technical foresight.

[0020] (4) Maintaining the advantages of system miniaturization and low cost: This invention uses a monochrome sensor, which does not require complex mechanical filter components, beam splitting prisms or multi-sensor beam splitting systems. It inherits the miniaturization and low cost characteristics of traditional time-division schemes, has a compact structure, and is easy to promote and apply in clinical and industrial scenarios.

[0021] (5) High illumination efficiency and suitable for low-light scenes: Multi-spectral mixed synchronous illumination is used for each exposure. Compared with single-spectral time-sharing illumination, the light energy utilization rate is significantly improved, and high-quality imaging can be obtained in low-light environments (such as internal cavities), reducing system power consumption.

[0022] (6) Flexible and scalable: The spectral composition and number of lighting sources can be expanded according to needs without changing the core hardware architecture, so as to adapt to imaging needs with a wider spectral range or more spectral channels.

[0023] (7) Excellent adaptive capability: The control and processing module can analyze scene brightness, contrast and motion in real time, and adaptively adjust the mixed lighting mode, exposure times and exposure time to adapt to various imaging conditions such as low illumination, high contrast and fast motion, reduce manual intervention and have a high degree of intelligence. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0025] Figure 1 A flowchart of the monochrome sensor hybrid coded illumination and image reconstruction method provided by the present invention; Figure 2 The image reconstruction algorithm flowchart provided by this invention; Figure 3 This is an overall architecture diagram of the endoscopic imaging system provided by the present invention; Figure 4 This is a schematic diagram of the structure of the hybrid coded lighting module provided by the present invention; Figure 5 Intensity ratio distribution diagrams for the three hybrid lighting modes provided by this invention; Figure 6 This image shows a comparison of the imaging effects of the present invention and a traditional time-sharing illumination scheme in a dynamic scene. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] To fundamentally address the color trailing problem in high-speed motion scenarios using traditional time-sharing illumination schemes, this invention proposes a method and corresponding imaging system based on a monochrome image sensor, which achieves color imaging through hybrid coded illumination and generalized linear model calculation and reconstruction. The system improves long-term color stability and environmental adaptability through a frame-level self-calibration mechanism, while maintaining its advantages of miniaturization, low cost, and high illumination efficiency. This makes it suitable for applications requiring high dynamic imaging performance and color stability, such as minimally invasive medical surgery navigation, internal cavity examination, and rapid industrial pipeline inspection.

[0028] Specifically, see Figure 1 This embodiment discloses a method for hybrid coded illumination and image reconstruction using a monochrome sensor, comprising the following steps: S1. Design at least three different RGB mixed lighting modes. In each mode, multiple light sources with different spectra are simultaneously illuminated and the intensity ratio of each spectrum is different. Construct the lighting encoding-color response matrix in the mixed lighting mode to realize the frame-level self-calibration mechanism. S2. The control lighting system performs at least three consecutive exposures according to a preset mixed lighting mode sequence. During each exposure, a complete grayscale image is acquired by the global shutter monochrome sensor in the corresponding mixed lighting mode. All exposures are completed under the same target state to ensure that there are no geometric position differences between the frames due to target movement. S3. Construct a generalized linear model based on the illumination coding-color response matrix, substitute the acquired multi-frame grayscale images into the generalized linear model to construct a system of linear equations, and reconstruct the color image of the target scene by solving the system of linear equations, and calculate the true intensity value of the RGB three channels of each pixel. S4. Post-process the reconstructed color image to output a high-quality color image in a standard format (such as sRGB format).

[0029] Furthermore, in S1, the light source in the RGB mixed illumination mode is a three-color LED light source consisting of red, green, and blue, and the number of spectral light sources can be expanded. The mixed illumination mode used in two adjacent exposures has a significant difference in the RGB intensity ratio to ensure good coding distinguishability; the design of the mixed illumination mode gives the coding matrix a good condition number, which can effectively improve the color reconstruction accuracy. Specifically, the acquisition of each frame image is based on a specific mixed illumination mode with a known intensity ratio, which constitutes the color coding of that frame image, and each frame image can be independently color-coded.

[0030] Furthermore, in S1, an illumination encoding-color response matrix for the mixed illumination mode is constructed. Specifically, using a standard color chart as the target, images are acquired under each RGB mixed illumination mode, and the spectral response coefficients of the sensor are determined through an optimization algorithm. , , A lighting coding-color response matrix is ​​established and stored to characterize the mapping relationship between the sensor grayscale response and the real color under different mixed lighting modes.

[0031] Furthermore, in S3, the optimization algorithm used to solve the linear equation system is any one of the following: Ordinary Least Squares (OLS), Weighted Least Squares (WLS), Ridge Regression, Principal Component Regression (PCR), or Singular Value Decomposition (SVD). The algorithm is flexibly selected based on noise conditions and the number of exposure frames to improve the robustness of color reconstruction. When the number of acquired image frames N > 3, optimization algorithms such as Weighted Least Squares and Regularized Least Squares are used to obtain more robust solutions.

[0032] For details, see Figure 2 Suppose N frames of images were captured (N≥3, usually equal to or greater than the number of lighting colors), and the intensity ratio of the RGB colors in the mixed lighting mode corresponding to each frame is: (in For each pixel location in the image, let its true RGB value be... In the i The grayscale value measured in the frame is Establish the following linear relationship:

[0033] In the formula: , , The spectral response coefficient of the sensor, This is the noise term.

[0034] Combine the N frames of observations into a system of equations:

[0035] Solve using the least squares method Thus, normalization is obtained. Value; when N>3, use regularized least squares estimation:

[0036] In the formula: These are weighting coefficients, allocated based on the signal-to-noise ratio; This is a regularization parameter used to prevent overfitting.

[0037] Furthermore, after reconstructing the color image in S3, for high-speed motion scenes, motion estimation algorithms such as optical flow can be combined to locally optimize the color image, thereby further improving the imaging quality.

[0038] Furthermore, the post-processing operations in S4 include one or more combinations of noise reduction, contrast adjustment, edge enhancement, gamma correction, and color space conversion, outputting standard format images that meet application requirements.

[0039] Furthermore, this embodiment also discloses an endoscopic imaging system that applies the monochromatic sensor hybrid coded illumination and image reconstruction method described above, such as... Figure 3 As shown, it includes: a hybrid coded illumination module, a monochrome image sensor module, a control and processing module, an image output module, and an endoscope probe, with each module integrated and packaged inside the endoscope probe; See Figure 4 The hybrid coded lighting module includes multiple light sources with different spectral characteristics (including at least components corresponding to the red, green, and blue spectra), an independent and controllable constant current drive circuit, and a light homogenizing component. The constant current drive circuit synchronously drives each light source and independently adjusts the luminous intensity of each spectral light source; the light homogenizing component is used to ensure the uniformity of the lighting light and avoid color reconstruction errors caused by uneven lighting. The monochrome image sensor module uses a global shutter monochrome CMOS sensor, which has a high dynamic range and low readout noise. It is used to quickly acquire grayscale images under mixed illumination modes and works with the global shutter to achieve single-frame synchronous acquisition. The control and processing module is based on a programmable logic device (FPGA) or an embedded processor and includes a lighting control unit, an image acquisition control unit, an image reconstruction and processing unit, and a self-calibration unit. The image output module is used to output the reconstructed color image to a display device or storage device.

[0040] Furthermore, in the control and processing module, the lighting control unit controls the switching of the hybrid lighting mode according to a preset sequence or adaptive strategy, the image acquisition control unit synchronously controls the exposure timing and data reading of the sensor to ensure accurate synchronization between exposure and lighting mode, the image reconstruction processing unit implements a variety of reconstruction algorithms under the framework of a generalized linear model, and the self-calibration unit implements a frame-level self-calibration mechanism and calibrates system parameters in real time.

[0041] Furthermore, the endoscope probe is designed to integrate the core components of the hybrid coded illumination module, monochrome image sensor module, and control and processing module into the probe, forming an integrated imaging unit. The probe has a compact structure with a diameter ranging from 1mm to 8mm, making it suitable for different application scenarios.

[0042] The technical solution of the present invention will be further described in detail below through the following examples.

[0043] (I) Basic Three-Exposure Hybrid Encoded Illumination Scheme

[0044] Example 1 provides a basic hybrid coded illumination and image reconstruction scheme that is applicable to most conventional endoscopic imaging scenarios such as gastroscopy and colonoscopy, balancing image quality and cost control.

[0045] 1. System Composition

[0046] Hybrid coding lighting module: It adopts red, green and blue LED light sources, and each LED is equipped with an independent controllable constant current drive circuit; it is equipped with a light uniform component to ensure the uniformity of lighting.

[0047] Monochrome image sensor module: A global shutter monochrome CMOS sensor is selected.

[0048] Control and processing module: Implemented based on FPGA, responsible for lighting mode switching control, sensor driving, image reconstruction calculation and basic frame-level self-calibration.

[0049] Endoscope probe: encapsulated in a medical-grade shell.

[0050] 2. Work Process

[0051] (1) System initialization and pre-calibration

[0052] Using a standard color chart as the calibration target, grayscale images were acquired under three preset mixed lighting modes. The spectral response coefficient of the sensor is solved by optimizing the algorithm; Establish and store the lighting encoding-color response matrix as the initial parameters for color reconstruction.

[0053] (2) Imaging process

[0054] See Figure 5 Three preset mixed lighting modes are provided, with the following RGB intensity ratios for each mode: Mode 1: ; Mode 2: ; Mode 3: .

[0055] The control and processing module drives the hybrid coded lighting module, switching cyclically in the order of mode 1, mode 2, and mode 3, with one exposure in each mode; A global shutter monochrome sensor simultaneously acquires three frames of grayscale images.

[0056] (3) Color reconstruction and output: The control and processing module solves a system of linear equations using the ordinary least squares method to calculate the RGB three-channel intensity value of each pixel. Post-processing is performed on the reconstructed color image; Output sRGB format color images through the image output module.

[0057] (4) Frame-level self-calibration: Periodically use stable regions in the image as references to fine-tune the spectral response coefficients and compensate for slight system drift.

[0058] 3. Performance Characteristics

[0059] Dynamic imaging performance: No color motion artifacts are observed when the motion speed is ≤5cm / s.

[0060] Color reproduction accuracy: Color reproduction error is less than 10%.

[0061] Applicable scenarios: Most routine medical endoscopic examinations such as gastroscopy and colonoscopy, as well as basic industrial endoscopic testing scenarios where cost is sensitive.

[0062] (II) Enhanced Four-Exposure Hybrid Encoded Illumination Scheme

[0063] Example 2 adds an exposure process to Example 1, optimizes the lighting mode design and reconstruction algorithm, and improves color reconstruction accuracy and noise robustness, making it suitable for professional scenarios with high image quality requirements.

[0064] 1. Enhanced features

[0065] Lighting mode optimization: Four mixed lighting modes are adopted to increase information redundancy and improve the anti-interference ability of the coding matrix.

[0066] Algorithm optimization: Weighted least squares (WLS) is used for color reconstruction, and weight coefficients are dynamically allocated according to the signal-to-noise ratio of each frame.

[0067] 2. Lighting pattern design

[0068] The RGB intensity ratios for the four hybrid lighting modes are designed as follows: Pattern 1: (0.7, 0.2, 0.1); Pattern 2: (0.2, 0.7, 0.1); Pattern 3: (0.1, 0.2, 0.7); Pattern 4: (0.4, 0.3, 0.3).

[0069] 3. High-speed motion imaging verification

[0070] like Figure 6 As shown, in a simulated high-speed motion test scenario, the traditional time-sharing illumination scheme produces obvious color ghosting; in the image output by the scheme in this embodiment, the edge color of the moving target is consistent with the body color, and there is no color separation phenomenon.

[0071] 4. Performance Improvement

[0072] Color reproduction error reduced to less than 5%; Noise levels are reduced by approximately 30%; It is suitable for scenarios such as minimally invasive surgical navigation and high-precision industrial defect detection.

[0073] (III) Examples of Multispectral Subdivision Illumination

[0074] Example 3 illustrates the scalability of the lighting spectrum. Beyond basic RGB tri-color lighting, a greater number of LED light sources with more subdivided spectra (e.g., subdividing the red band into two different wavelengths of red LEDs) can be used to create mixed lighting patterns with more spectral channels. The generalized linear model in the reconstruction algorithm is correspondingly extended to more dimensions to solve for more spectral components, based on the same principle as in Example 1. This extension further enhances the potential accuracy of color reproduction and spectral resolution.

[0075] In summary, the core differences and technical advantages of the technical solution in this embodiment compared to the prior art are summarized as follows: (1) Different color acquisition mechanisms: Traditional time-sharing lighting schemes use different spectral channels for time-separated acquisition, and color trailing is caused by the time difference of acquisition of each channel; while the present invention uses single-frame spectral encoding for complete acquisition, and each frame image contains full color encoding information of the target scene. There is no time difference in color information, and even if the target moves at high speed between frames, there is no color misalignment within a single frame.

[0076] (2) Different system calibration mechanisms: Traditional solutions use a one-time pre-calibration with fixed calibration parameters, which cannot compensate for drift during long-term operation of the system; while the present invention has frame-level self-calibration capability, which uses the known illumination code of each frame and combines scene reference information to back-calculate and calibrate system parameters in real time, thereby improving long-term color stability.

[0077] (3) Different algorithm compatibility: The generalized linear model framework proposed in this invention is compatible with a variety of existing and future linear reconstruction algorithms, expanding the scope of patent protection and enhancing the foresight and application flexibility of the technology.

[0078] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0079] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for hybrid coded illumination and image reconstruction using a monochrome sensor, characterized in that, Includes the following steps: S1. Design at least three different RGB mixed lighting modes, in which multiple light sources of different spectra are simultaneously illuminated and the intensity ratio of each spectrum is different, and construct the lighting encoding-color response matrix of the mixed lighting mode; S2. The control lighting system performs at least three consecutive exposures according to a preset mixed lighting mode sequence. During each exposure, a complete grayscale image is acquired by the global shutter monochrome sensor in the corresponding mixed lighting mode, and all exposures are completed under the same target state. S3. Construct a generalized linear model based on the illumination coding-color response matrix, substitute the acquired multi-frame grayscale images into the generalized linear model to construct a system of linear equations, and reconstruct the color image of the target scene by solving the system of linear equations, and calculate the true intensity value of the RGB three channels of each pixel. S4. Post-process the reconstructed color image and output a standard format color image.

2. The method for hybrid coded illumination and image reconstruction using a monochrome sensor according to claim 1, characterized in that, In S1, the light source in the RGB mixed lighting mode is a red, green, and blue LED light source, and the number of spectral light sources can be expanded.

3. The method for hybrid coded illumination and image reconstruction using a monochrome sensor according to claim 1, characterized in that, In S1, an illumination coding-color response matrix under mixed lighting modes is constructed. Specifically, using a standard color chart as the target, images are acquired under each RGB mixed lighting mode. The spectral response coefficients of the sensor are determined through an optimization algorithm. An illumination coding-color response matrix that characterizes the mapping relationship between the sensor's grayscale response and the real color under different mixed lighting modes is established and stored.

4. The method for hybrid coded illumination and image reconstruction using a monochrome sensor according to claim 1, characterized in that, In S3, the optimization algorithm used to solve the linear equation system is any one of the following: least squares method, weighted least squares method, ridge regression, principal component regression or singular value decomposition method, which can be flexibly selected according to noise conditions and exposure frame number.

5. The method for hybrid coded illumination and image reconstruction using a monochrome sensor according to claim 1, characterized in that, After reconstructing the color image in S3, the color image is locally optimized using optical flow.

6. The method for hybrid coded illumination and image reconstruction using a monochrome sensor according to claim 1, characterized in that, Post-processing operations in S4 include one or more combinations of noise reduction, contrast adjustment, edge enhancement, gamma correction, and color space conversion.

7. An endoscopic imaging system, characterized in that, The monochrome sensor hybrid coded illumination and image reconstruction method according to any one of claims 1-6 includes: a hybrid coded illumination module, a monochrome image sensor module, a control and processing module, an image output module, and an endoscope probe, with each module integrated and packaged inside the endoscope probe; The hybrid coded lighting module includes multiple light sources with different spectral characteristics, an independent and controllable constant current drive circuit, and a light homogenizing component. The constant current drive circuit synchronously drives each light source and independently adjusts the luminous intensity of each spectral light source. The monochrome image sensor module uses a global shutter monochrome CMOS sensor to acquire grayscale images under mixed illumination modes; The control and processing module is based on a programmable logic device or an embedded processor and includes a lighting control unit, an image acquisition control unit, an image reconstruction and processing unit, and a self-calibration unit. The image output module is used to output the reconstructed color image to a display device or storage device.

8. An endoscopic imaging system according to claim 7, characterized in that, In the control and processing module, the lighting control unit controls the switching of the mixed lighting mode according to a preset sequence or adaptive strategy, the image acquisition control unit synchronously controls the exposure timing and data reading of the sensor, the image reconstruction processing unit implements a variety of reconstruction algorithms under the framework of a generalized linear model, and the self-calibration unit implements a frame-level self-calibration mechanism and calibrates the system parameters in real time.

9. An endoscopic imaging system according to claim 7, characterized in that, The endoscope probe is a compact, integrated imaging unit with a diameter ranging from 1mm to 8mm.

Images