A hemoglobin emphasizing device and method for an endoscope

Abstract

The application relates to the technical field of medical image processing, in particular to a hemoglobin emphasis device and method for an endoscope. The device comprises a light source part; a shooting part used for collecting reflected light signals of a target area under the multispectral illumination and outputting original Bayer format images or RGB color images; an image processing part used for receiving and processing the outputted Bayer format images or RGB color images, which at least comprises: a spectrum reconstruction module used for calculating and reconstructing single-band image signals by using a spectrum reconstruction model; a hemoglobin concentration calculation module configured to calculate relative hemoglobin concentration based on the image signals; and an image enhancement module configured to perform pixel-level gain adjustment on the original color images or the processed images, convert to a set color space for contrast mapping, and finally output hemoglobin emphasized images. The outputted images can significantly improve the contrast between blood vessels and surrounding tissues.

Description

Technical Field

[0001] This application relates to the field of medical image processing technology, specifically to a hemoglobin emphasis device and method for endoscopes. Background Technology

[0002] In endoscopic diagnosis and treatment, clear identification of submucosal vessels is crucial for early cancer diagnosis and surgical navigation. However, due to the scattering and absorption of light by biological tissues, as well as the non-uniformity of lighting conditions, the contrast between blood vessels and background tissues is often low, making it difficult to observe the morphology of blood vessels.

[0003] In existing technologies, techniques such as narrow-band imaging (NBI) enhance the visibility of blood vessels by using narrow-band light of specific wavelengths. However, these techniques typically rely on special optical filters, which are costly and lack flexibility, making them difficult to implement on ordinary endoscopes. Furthermore, traditional image enhancement algorithms often amplify noise while enhancing blood vessels, or cause image color distortion, affecting the doctor's clinical judgment.

[0004] Therefore, there is an urgent need in the field for a method and device that can effectively and adaptively emphasize hemoglobin signals and highlight vascular structures using ordinary endoscopic hardware and advanced image processing algorithms. Summary of the Invention

[0005] This application is mainly about a hemoglobin emphasis device and method for endoscopes, which solves the problems of the prior art.

[0006] In a first aspect, this application provides a hemoglobin emphasis device for an endoscope, comprising:

[0007] The light source unit, used to provide multispectral illumination to the target area of ​​biological tissue, consists of multiple laser diodes or light-emitting diodes with different center wavelengths;

[0008] The imaging unit is used to acquire the reflected light signal of the target area under the multispectral illumination and output the original Bayer format image or RGB color image;

[0009] An image processing unit, electrically connected to the imaging unit, is used to receive and process output Bayer format images or RGB color images, and includes at least:

[0010] The spectral reconstruction module is configured to take the original image acquired by the imaging unit as input and use the spectral reconstruction model to calculate and reconstruct three single-band image signals with center wavelengths of blue, green and red, respectively, which are denoted as the first image signal, the second image signal and the third image signal.

[0011] The hemoglobin concentration calculation module is configured to calculate the relative hemoglobin concentration of the first image signal, the second image signal, and the third image signal, with the third image signal as the reference, relative to the first image signal and the second image signal respectively.

[0012] as well as,

[0013] The image enhancement module is configured to perform pixel-level gain adjustment on the original color image or the processed image based on the calculated relative hemoglobin concentration, and convert it to a set color space for contrast mapping, ultimately outputting a hemoglobin-enhanced image.

[0014] Based on the above technical solutions, the spectral reconstruction model is either a linear model or a nonlinear model.

[0015] Based on the above technical solutions, the spectral reconstruction model is a three-layer fully connected neural network with ReLU activation function or a single-layer linear mapping network with least squares estimation.

[0016] Based on the above technical solution, it also includes a display unit, which is used to display the output hemoglobin-enhanced image.

[0017] Secondly, this application provides a hemoglobin emphasis method for endoscopy, which includes the following steps:

[0018] S1

[0019] Illuminate the target area of ​​biological tissue using multiple light sources with different center wavelengths, and among the multiple light sources with different center wavelengths, at least one light source with a center wavelength of blue, green and red is included respectively;

[0020] S2

[0021] Raw image signals obtained from the target region of biological tissue;

[0022] S3

[0023] Based on the original image signal, a spectral reconstruction model was used to reconstruct three single-band image signals with predetermined center wavelengths: blue, green, and red. These are denoted as the first image signal. Second image signal and third image signal ;

[0024] S4

[0025] The first relative hemoglobin concentration C1 is calculated from the third image signal and the first image signal, and its calculation formula (1) is as follows:

[0026] (1)

[0027] The second relative hemoglobin concentration C2 is calculated from the third image signal and the second image signal, and its calculation formula (2) is as follows:

[0028] (2)

[0029] In equations (1) and (2), g1(·) and g2(·) are both linear or nonlinear functions used to map the light intensity ratio to a concentration value.

[0030] S5

[0031] Based on the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, the gain coefficients of the red, green, and blue channels of the original color image are calculated respectively, using the following formulas:

[0032] (3)

[0033] In equation (3), This represents the gain coefficient of the red channel in the original color image. This represents the gain coefficient of the green channel in the original color image. This represents the gain coefficient of the blue channel in the original color image; , , These are the conversion functions for the red, green, and blue channels, respectively. These conversion functions can be linear functions, exponential functions, lookup table functions, or other nonlinear functions.

[0034] The calculated gain coefficients for the three channels are multiplied by the corresponding pixel values ​​of the original color image to obtain the preliminarily enhanced image.

[0035] S6

[0036] The initially enhanced image is converted from the RGB color space to the CIE Lab color space;

[0037] In the Lab color space, the piecewise linear or nonlinear mapping of the a component representing the red-green spectrum is performed to obtain the mapped a' component.

[0038] The mapped a' component is recombined with the L and b components in the CIE Lab color space and converted back to the RGB color space to obtain the final hemoglobin-enhanced image.

[0039] Based on the above method, in step S4,

[0040] Set an average image brightness threshold T, and extract pixel regions from the first image signal, the second image signal and the third image signal whose average image brightness is higher than the average image brightness threshold T as the extraction region;

[0041] The signal values ​​of the extraction regions corresponding to the first image signal, the second image signal, and the third image signal are extracted respectively to calculate the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2.

[0042] Based on the above method, in step S4, both g1(·) and g2(·) are logarithmic functions, that is:

[0043] (4)

[0044] (5)

[0045] In equations (4) and (5), k1 and k2 are constants used to control the range of values ​​of the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, so that their values ​​are between (-1, 1).

[0046] Based on the above method, in step S5, , , Both are exponential functions, that is:

[0047] (6)

[0048] In equation (6), C0 is the reference value, and Kr, Kg, and Kb are normal numbers used to control the range of gain coefficients so that the gain coefficients of the red, green, and blue channels are between (0 and 2).

[0049] Based on the above method, in step S6, the a component adopts a nonlinear mapping, and the specific formula is as follows:

[0050] (7)

[0051] In equation (7), α, γ and β are all mapping parameters.

[0052] Based on the above method, in step S6, component a adopts a piecewise linear mapping, the specific formula of which is:

[0053] (8)

[0054] in,

[0055] (9)

[0056] In equations (8) and (9), and These are all adjustable parameters corresponding to different organs in the human body.

[0057] This application provides a hemoglobin enhancement device and method for endoscopy, aiming to improve the visualization of blood vessels and assist doctors in making more accurate diagnoses. The device does not require the integration of special optical filters into the endoscope, and can achieve narrowband imaging effects using ordinary light sources and multispectral reconstruction algorithms, reducing system cost and complexity. By calculating the concentrations of two relative hemoglobins and adjusting the multi-channel adaptive gain accordingly, the device can accurately enhance blood vessel signals while effectively suppressing background noise. During processing, nonlinear mapping is performed in the color space, and an adjustable parameter set is provided, enabling the method to adapt to the imaging needs of different organs, individuals, and lighting conditions, achieving stable and reliable enhancement effects. The final output image can significantly improve the contrast between blood vessels and surrounding tissues, helping doctors detect more subtle lesions, improve the detection rate of early cancer, and enhance the precision of surgical procedures. Attached Figure Description

[0058] The accompanying drawings, which are included to provide a further understanding of the embodiments of this application and form part of this application, do not constitute a limitation on the embodiments of this application. In the drawings:

[0059] Figure 1 A schematic diagram of a hemoglobin emphasis device used in endoscopy;

[0060] Figure 2 This is a schematic diagram of the image data effect in a specific embodiment;

[0061] Figure 3 This is a schematic diagram of the nonlinear mapping of the a-component in the Lab space in a specific embodiment one;

[0062] Figure 4 This is a schematic diagram of piecewise linear mapping of component a in Lab space in specific embodiment two. Detailed Implementation

[0063] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0064] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0065] like Figure 1 As shown, the first embodiment of the present invention provides a hemoglobin emphasis device for an endoscope, comprising:

[0066] The light source unit, used to provide multispectral illumination to the target area of ​​biological tissue, consists of multiple laser diodes or light-emitting diodes with different center wavelengths;

[0067] The imaging unit is used to acquire the reflected light signal of the target area under the multispectral illumination and output the original Bayer format image or RGB color image;

[0068] An image processing unit, electrically connected to the imaging unit, is used to receive and process output Bayer format images or RGB color images, and includes at least:

[0069] The spectral reconstruction module is configured to take the original image acquired by the imaging unit as input and use the spectral reconstruction model to calculate and reconstruct three single-band image signals with center wavelengths of blue, green and red, respectively, which are denoted as the first image signal, the second image signal and the third image signal.

[0070] The hemoglobin concentration calculation module is configured to calculate the relative hemoglobin concentration of the first image signal, the second image signal, and the third image signal, with the third image signal as the reference, respectively, relative to the first image signal and the second image signal.

[0071] as well as,

[0072] The image enhancement module is configured to perform pixel-level gain adjustment on the original color image or the processed image based on the calculated relative hemoglobin concentration, and convert it to a set color space for contrast mapping, ultimately outputting a hemoglobin-enhanced image.

[0073] In this embodiment, the light source can be the endoscope's own optical system, the imaging unit can be the image sensor (such as CMOS or CCD) at the front end of the endoscope, and the image processing unit can be an FPGA embedded in the endoscope's image processor and a software module that works in conjunction with it.

[0074] As a specific implementation method, the spectral reconstruction model is a linear model or a nonlinear model.

[0075] As a specific implementation, the spectral reconstruction model is a three-layer fully connected neural network with ReLU activation function or a single-layer linear mapping network with least squares estimation.

[0076] In a specific embodiment, a display unit is also included for displaying the output hemoglobin-enhanced image. Specifically, the display unit can be a medical monitor connected to the endoscope signal.

[0077] A second embodiment of the present invention provides a hemoglobin emphasis method for endoscopy, comprising the following steps:

[0078] S1

[0079] Illuminate the target area of ​​biological tissue using multiple light sources with different center wavelengths, and among the multiple light sources with different center wavelengths, at least one light source with a center wavelength of blue, green and red is included respectively;

[0080] S2

[0081] Raw image signals obtained from the target region of biological tissue;

[0082] S3

[0083] Based on the original image signal, a spectral reconstruction model was used to reconstruct three single-band image signals with predetermined center wavelengths: blue, green, and red. These are denoted as the first image signal. Second image signal and third image signal ;

[0084] S4

[0085] The first relative hemoglobin concentration C1 is calculated from the third image signal and the first image signal, and its calculation formula (1) is as follows:

[0086] (1)

[0087] The second relative hemoglobin concentration C2 is calculated from the third image signal and the second image signal, and its calculation formula (2) is as follows:

[0088] (2)

[0089] In equations (1) and (2), g1(·) and g2(·) are both linear or nonlinear functions used to map the light intensity ratio to a concentration value.

[0090] S5

[0091] Based on the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, the gain coefficients of the red, green, and blue channels of the original color image are calculated respectively, using the following formulas:

[0092] (3)

[0093] In equation (3), This represents the gain coefficient of the red channel in the original color image. This represents the gain coefficient of the green channel in the original color image. This represents the gain coefficient of the blue channel in the original color image; , , These are the conversion functions for the red, green, and blue channels, respectively. These conversion functions can be linear functions, exponential functions, lookup table functions, or other nonlinear functions.

[0094] The calculated gain coefficients for the three channels are multiplied by the corresponding pixel values ​​of the original color image to obtain the preliminarily enhanced image.

[0095] S6

[0096] The initially enhanced image is converted from the RGB color space to the CIE Lab color space;

[0097] In the Lab color space, the piecewise linear or nonlinear mapping of the a component representing the red-green spectrum is performed to obtain the mapped a' component.

[0098] The mapped a' component is recombined with the L and b components in the CIE Lab color space and converted back to the RGB color space to obtain the final hemoglobin-enhanced image.

[0099] As a specific implementation method, in step S4,

[0100] Set an average image brightness threshold T, and extract pixel regions from the first image signal, the second image signal and the third image signal whose average image brightness is higher than the average image brightness threshold T as the extraction region;

[0101] The signal values ​​of the extraction regions corresponding to the first image signal, the second image signal, and the third image signal are extracted respectively to calculate the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2.

[0102] In a specific implementation, in step S4, both g1(·) and g2(·) are logarithmic functions, that is:

[0103] (4)

[0104] (5)

[0105] In equations (4) and (5), k1 and k2 are constants used to control the range of values ​​of the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, so that their values ​​are between (-1, 1).

[0106] As one specific implementation method, in step S5, , , Both are exponential functions, that is:

[0107] (6)

[0108] In equation (6), C0 is the reference value, and Kr, Kg, and Kb are normal numbers used to control the range of gain coefficients so that the gain coefficients of the red, green, and blue channels are between (0 and 2).

[0109] As one specific implementation method, in step S6, the a component adopts a nonlinear mapping, and the specific formula is as follows:

[0110] (7)

[0111] In equation (7), α, γ and β are all mapping parameters.

[0112] In a specific implementation, in step S6, component a adopts a piecewise linear mapping, and the specific formula is as follows:

[0113] (8)

[0114] in,

[0115] (9)

[0116] In equations (8) and (9), and These are all adjustable parameters corresponding to different organs of the human body. It should be noted that, depending on the final image contrast, these parameters can be flexibly selected and adjusted according to the different organs of the human body. and The range and value of the adjustable parameters are not further limited in this embodiment, as long as the final output image can significantly improve the contrast between blood vessels and surrounding tissues.

[0117] In summary, the aforementioned hemoglobin enhancement device and method for endoscopy aims to improve the visualization of blood vessels and assist doctors in making more accurate diagnoses. The device eliminates the need for special optical filters integrated into the endoscope, achieving narrowband imaging using ordinary light sources and multispectral reconstruction algorithms, thus reducing system cost and complexity. By calculating two relative hemoglobin concentrations and performing multi-channel adaptive gain adjustment accordingly, it can accurately enhance vascular signals while effectively suppressing background noise. During processing, nonlinear mapping in the color space and the provision of an adjustable parameter set allow the method to adapt to the imaging needs of different organs, individuals, and lighting conditions, achieving stable and reliable enhancement effects. The final output image significantly improves the contrast between blood vessels and surrounding tissues, helping doctors detect more subtle lesions, increasing the detection rate of early cancer, and improving the precision of surgical procedures.

[0118] Below, in order to better understand and implement the technical solution of the present invention, the present invention will be explained and described in detail with reference to specific embodiments. Specific Implementation Example 1:

[0120] A hemoglobin emphasis device for endoscopes includes a light source, an imaging unit, and an image processing unit. The light source consists of LEDs with center wavelengths of 450nm, 550nm, and 660nm; the imaging unit is a 4K CMOS image sensor; and the image processing unit consists of an FPGA embedded in the endoscope image processor and a software module that works in conjunction with it.

[0121] When applying this technique, hemoglobin emphasis in the endoscope is achieved using the following methods:

[0122] A method for hemoglobin emphasis in endoscopy, comprising the following steps:

[0123] S1

[0124] The target area of ​​the biological tissue was illuminated using LED light sources with center wavelengths of 450nm, 550nm and 660nm respectively.

[0125] S2

[0126] Raw image signals obtained from the target region of biological tissue;

[0127] S3

[0128] Based on the original image signal, three single-band image signals with center wavelengths of 450nm, 550nm and 660nm were reconstructed using a spectral reconstruction model, and were denoted as the first image signal, the second image signal and the third image signal, respectively.

[0129] The spectral reconstruction model uses a pre-trained three-layer fully connected neural network (containing the ReLU activation function) to map the original image pixel values ​​to... The intensity value, such as Figure 2 As shown in the first step;

[0130] S4

[0131] The first relative hemoglobin concentration C1 is calculated from the third image signal and the first image signal, and the calculation formula is as follows:

[0132] (4)

[0133] The second relative hemoglobin concentration C2 is calculated from the third and second image signals using the following formula:

[0134] (5)

[0135] In equations (4) and (5), g1(·) and g2(·) are both logarithmic functions; k1 and k2 are both constants, where k1=0.33 and k2=0.33.

[0136] In a specific implementation, an average image brightness threshold T is set, and pixel regions in the first, second, and third image signals whose average image brightness is higher than the average image brightness threshold T are extracted as extraction regions. The signal values ​​of the extraction regions corresponding to the first, second, and third image signals are extracted respectively to calculate the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2.

[0137] Specifically, the average image brightness threshold T is 30% of the average image brightness.

[0138] S5

[0139] Based on the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, the gain coefficients of the red, green, and blue channels of the original color image are calculated respectively, using the following formulas:

[0140] (6)

[0141] In equation (6), This represents the gain coefficient of the red channel in the original color image. This represents the gain coefficient of the green channel in the original color image. This represents the gain coefficient of the blue channel in the original color image; C0 is the baseline value, and Kr, Kg, and Kb are positive constants used to control the range of gain coefficient values, ensuring that the gain coefficients of the red, green, and blue channels are between 0 and 2. Specifically, C0=1, Kr = 0.2, Kg=0.3, Kb=0.3.

[0142] The calculated gain coefficients for the three channels are multiplied by the corresponding pixel values ​​of the original color image to obtain the preliminarily enhanced image, such as... Figure 2 The second step is shown; the specific formula is as follows:

[0143] (10)

[0144] Where R, G, B are the original pixel values ​​of the image, and R', G', B' are the enhanced pixel values.

[0145] S6

[0146] The initially enhanced image is converted from the RGB color space to the CIE Lab color space;

[0147] In the Lab color space, the piecewise linear or nonlinear mapping of the 'a' component, which represents the red-green spectrum, yields the mapped 'a' component. The mapping curve is shown below. Figure 3 As shown;

[0148] The 'a' component employs a nonlinear mapping, with the specific formula as follows:

[0149] (7)

[0150] In equation (7), α, γ and β are all mapping parameters, specifically α=1.2, γ=0.85 and β=0.01.

[0151] The mapped a' component is recombine with the L and b components in the CIE Lab color space and converted back to the RGB color space to obtain the final hemoglobin-emphasized image, as shown below. Figure 2 The third step is shown. Specific Implementation Example 2:

[0153] A hemoglobin emphasis device for endoscopes includes a light source, an imaging unit, and an image processing unit. The light source consists of LEDs with center wavelengths of 450nm, 550nm, and 660nm; the imaging unit is a 4K CMOS image sensor; and the image processing unit consists of an FPGA embedded in the endoscope image processor and a software module that works in conjunction with it.

[0154] When applying this technique, hemoglobin emphasis in the endoscope is achieved using the following methods:

[0155] A method for hemoglobin emphasis in endoscopy, comprising the following steps:

[0156] S1

[0157] The target area of ​​the biological tissue was illuminated using LED light sources with center wavelengths of 450nm, 550nm and 660nm respectively.

[0158] S2

[0159] Raw image signals obtained from the target region of biological tissue;

[0160] S3

[0161] Based on the original image signal, three single-band image signals with center wavelengths of 450nm, 550nm and 660nm were reconstructed using a spectral reconstruction model, and were denoted as the first image signal, the second image signal and the third image signal, respectively.

[0162] The spectral reconstruction model uses a pre-trained three-layer fully connected neural network (containing the ReLU activation function) to map the original image pixel values ​​to... The intensity value, such as Figure 2 As shown in the first step;

[0163] S4

[0164] The first relative hemoglobin concentration C1 is calculated from the third image signal and the first image signal, and its calculation formula (1) is as follows:

[0165] (4)

[0166] The second relative hemoglobin concentration C2 is calculated from the third image signal and the second image signal, and its calculation formula (2) is as follows:

[0167] (5)

[0168] In equations (4) and (5), g1(·) and g2(·) are both logarithmic functions; k1 and k2 are both constants, where k1=0.33 and k2=0.33.

[0169] In a specific implementation, an average image brightness threshold T is set, and pixel regions in the first, second, and third image signals whose average image brightness is higher than the average image brightness threshold T are extracted as extraction regions. The signal values ​​of the extraction regions corresponding to the first, second, and third image signals are extracted respectively to calculate the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2.

[0170] Specifically, the average image brightness threshold T is 30% of the average image brightness.

[0171] S5

[0172] Based on the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, the gain coefficients of the red, green, and blue channels of the original color image are calculated respectively, using the following formulas:

[0173] (6)

[0174] In equation (6), This represents the gain coefficient of the red channel in the original color image. This represents the gain coefficient of the green channel in the original color image. This represents the gain coefficient of the blue channel in the original color image; C0 is the baseline value, and Kr, Kg, and Kb are positive constants used to control the range of gain coefficient values, ensuring that the gain coefficients of the red, green, and blue channels are between 0 and 2. Specifically, C0=1, Kr = 0.2, Kg=0.3, Kb=0.3.

[0175] The calculated gain coefficients for the three channels are multiplied by the corresponding pixel values ​​of the original color image to obtain the preliminarily enhanced image, such as... Figure 2 The second step is shown; the specific formula is as follows:

[0176] (10)

[0177] Where R, G, B are the original pixel values ​​of the image, and R', G', B' are the enhanced pixel values.

[0178] S6

[0179] The initially enhanced image is converted from the RGB color space to the CIE Lab color space;

[0180] In the Lab color space, the piecewise linear or nonlinear mapping of the 'a' component, which represents the red-green spectrum, yields the mapped 'a' component. The mapping curve is shown below. Figure 4 As shown;

[0181] The 'a' component employs a piecewise linear mapping, with the specific formula as follows:

[0182] (8)

[0183] in,

[0184] (9)

[0185] In equations (8) and (9), and These are all adjustable parameters corresponding to different organs in the human body. Specifically, when the human organ is the oral cavity, , .

[0186] The mapped a' component is recombine with the L and b components in the CIE Lab color space and converted back to the RGB color space to obtain the final hemoglobin-emphasized image, as shown below. Figure 2 The third step is shown.

[0187] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

[0188] Those skilled in the art will understand that embodiments of the present invention can be provided as methods or systems. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0189] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0190] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0191] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the foregoing claims.

[0192] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for hemoglobin emphasis in endoscopy, characterized in that, Includes the following steps: S1 Illuminate the target area of ​​biological tissue using multiple light sources with different center wavelengths, and among the multiple light sources with different center wavelengths, at least one light source with a center wavelength of blue, green and red is included respectively; S2 Raw image signals obtained from the target region of biological tissue; S3 Based on the original image signal, a spectral reconstruction model was used to reconstruct three single-band image signals with predetermined center wavelengths: blue, green, and red. These are denoted as the first image signal. Second image signal and third image signal ; S4 The first relative hemoglobin concentration C1 is calculated from the third image signal and the first image signal, and its calculation formula (1) is as follows: （1） The second relative hemoglobin concentration C2 is calculated from the third image signal and the second image signal, and its calculation formula (2) is as follows: （2） In equations (1) and (2), g1(·) and g2(·) are both linear or nonlinear functions used to map the light intensity ratio to a concentration value. S5 Based on the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, the gain coefficients of the red, green, and blue channels of the original color image are calculated respectively, using the following formulas: （3） In equation (3), This represents the gain coefficient of the red channel in the original color image. This represents the gain coefficient of the green channel in the original color image. This represents the gain coefficient of the blue channel in the original color image; , , These are the conversion functions for the red, green, and blue channels, respectively. These conversion functions can be linear functions, exponential functions, lookup table functions, or other nonlinear functions. The calculated gain coefficients for the three channels are multiplied by the corresponding pixel values ​​of the original color image to obtain the preliminarily enhanced image. S6 The initially enhanced image is converted from the RGB color space to the CIE Lab color space; In the Lab color space, the piecewise linear or nonlinear mapping of the a component representing the red-green spectrum is performed to obtain the mapped a' component. The mapped a' component is recombined with the L and b components in the CIE Lab color space and converted back to the RGB color space to obtain the final hemoglobin-enhanced image.

2. The method for hemoglobin emphasis in endoscopy according to claim 1, characterized in that, In step S4 Set an average image brightness threshold T, and extract pixel regions from the first image signal, the second image signal and the third image signal whose average image brightness is higher than the average image brightness threshold T as the extraction region; The signal values ​​of the extraction regions corresponding to the first image signal, the second image signal, and the third image signal are extracted respectively to calculate the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2.

3. The method for hemoglobin emphasis in endoscopy according to claim 1, characterized in that, In step S4, both g1(·) and g2(·) are logarithmic functions, that is: （4） （5） In equations (4) and (5), k1 and k2 are constants used to control the range of values ​​of the first relative hemoglobin concentration C1 and the second relative hemoglobin concentration C2, so that their values ​​are between (-1, 1).

4. The method for hemoglobin emphasis in endoscopy according to claim 1, characterized in that, In step S5 , , Both are exponential functions, that is: （6） In equation (6), C0 is the reference value, and Kr, Kg, and Kb are normal numbers used to control the range of gain coefficients so that the gain coefficients of the red, green, and blue channels are between (0 and 2).

5. The method for hemoglobin emphasis in endoscopy according to claim 1, characterized in that, In step S6, the a component adopts a nonlinear mapping, and the specific formula is as follows: （7） In equation (7), α, γ and β are all mapping parameters.

6. The method for hemoglobin emphasis in endoscopy according to claim 1, characterized in that, In step S6, component a adopts a piecewise linear mapping, the specific formula of which is: （8） in, （9） In equations (8) and (9), and These are all adjustable parameters corresponding to different organs in the human body.

7. A hemoglobin emphasis device for an endoscope, used to implement the hemoglobin emphasis method according to any one of claims 1-6, characterized in that, include: The light source unit, used to provide multispectral illumination to the target area of ​​biological tissue, consists of multiple laser diodes or light-emitting diodes with different center wavelengths; The imaging unit is used to acquire the reflected light signal of the target area under the multispectral illumination and output the original Bayer format image or RGB color image; An image processing unit, electrically connected to the imaging unit, is used to receive and process output Bayer format images or RGB color images, and includes at least: The spectral reconstruction module is configured to take the original image acquired by the imaging unit as input and use the spectral reconstruction model to calculate and reconstruct three single-band image signals with center wavelengths of blue, green and red, respectively, which are denoted as the first image signal, the second image signal and the third image signal. The hemoglobin concentration calculation module is configured to calculate the relative hemoglobin concentration of the first image signal, the second image signal, and the third image signal, with the third image signal as the reference, relative to the first image signal and the second image signal respectively. as well as, The image enhancement module is configured to perform pixel-level gain adjustment on the original color image or the processed image based on the calculated relative hemoglobin concentration, and convert it to a set color space for contrast mapping, ultimately outputting a hemoglobin-enhanced image.

8. The hemoglobin emphasis device according to claim 7, characterized in that, The spectral reconstruction model can be a linear model or a nonlinear model.

9. The hemoglobin emphasis device according to claim 8, characterized in that, The spectral reconstruction model is a three-layer fully connected neural network with ReLU activation function or a single-layer linear mapping network with least squares estimation.

10. The hemoglobin emphasis device according to claim 7, characterized in that, It also includes a display unit, which is used to display the output hemoglobin-enhanced image.

Images