Endoscopic image forming method and endoscope

Abstract

The present invention relates to the field of medical endoscopic imaging technology, and discloses an endoscopic image imaging method and an endoscope, wherein the imaging method comprises: when the optical power of a first narrow-band light source changes, obtaining an original captured image of a target object captured by an imaging unit; decomposing a standard first color channel of the imaging unit corresponding to the standard optical power in the original first color channel; fusing the standard first color channel, the original second color channel, and the original third color channel to obtain a corrected image; respectively extracting brightness information representing brightness and details in the original captured image and color information in the corrected image, and fusing the brightness information and the color information to obtain an endoscopic imaging image. The adjusted endoscopic imaging image retains the detail enhancement brought about by the change in the ratio of the first narrow-band light source, while maintaining the color expression of the preset standard light imaging.

Description

Technical Field

[0001] The present invention relates to the technical field of medical endoscopic imaging, and in particular to an endoscopic image forming method and an endoscope. Background Art

[0002] With the vigorous development of endoscopy applications, doctors' demand for high-quality endoscopy imaging is also increasing. In endoscopy application scenarios, especially during the inspection, diagnosis and treatment of digestive endoscopy (gastroenteroscopy), doctors usually use the white light imaging mode that is consistent with human vision for routine operations. When further exploration of lesions is required, they will switch to narrow-band light imaging mode.

[0003] Narrowband imaging is an optical imaging technology that uses one or several narrow-band spectrum light sources as illumination to enhance the visibility of mucosal blood vessels, lesions, inflammation and other parts; common narrowband imaging modes include narrow-band imaging (NBI), blue laser imaging (BLI), linked color imaging (LCI), etc.; because narrowband imaging only covers part of the visible spectrum or changes the proportion of the white light spectrum, the color of the image is significantly different from that of white light imaging.

[0004] Since different parts of the human body cavity have different absorption and reflection characteristics for blue light and blue-violet light, increasing or decreasing the proportion of blue light and blue-violet light in different parts can effectively improve the distinction between mucosa, blood vessels and tissues. For example, the oral mucosal epithelium, which is mainly composed of stratified squamous epithelium, the gastric mucosal surface epithelium, which is mainly composed of columnar epithelial cells, and the duodenum, which is excreted by the liver and the bile duct to a large amount of bilirubin, have obvious differences in the absorption and reflection characteristics of blue light and blue-violet light. Appropriate increase or decrease in the proportion of blue light and blue-violet light will significantly improve the contrast and detail information of the corresponding parts.

[0005] However, for white light illumination, the corresponding power ratio of each LED light source at a given white light brightness level is fixed. If the ratio of blue light or blue-violet light is changed, it will obviously affect the light quality parameters such as color rendering index, color temperature, and chromaticity coordinates, which will in turn reduce the color reproduction ability of the camera system. At the same time, since each LED light source has many levels, there will be more combinations of levels of multiple light sources. Changes in each combination may lead to changes in the color rendering of the light source. It is impossible to match corresponding image parameters such as white balance, CCM, etc. for each combination.

[0006] Therefore, how to maintain the color expression of white light imaging while retaining the detail enhancement brought by changing the proportion of narrow-band light sources has become an urgent problem to be solved. Summary of the invention

[0007] In view of this, the present invention provides an endoscopic image imaging method, apparatus, computer equipment, storage medium and endoscope to solve the technical problem of how to maintain the color expression of white light imaging while retaining the detail enhancement brought by changing the ratio of narrow-band light sources.

[0008] According to a first aspect, the present invention provides an endoscopic image imaging method, the endoscope comprising an illumination unit, an imaging unit and an image processing unit, wherein the illumination unit comprises at least one first narrow-band light source of at least one first band corresponding to a blue channel in an RGB image collected by the imaging unit and a plurality of second light sources of a plurality of second bands corresponding to a red channel and a green channel, the first narrow-band light source and the second light source being able to synthesize a preset standard light under corresponding standard light power, and the imaging unit being able to collect a standard image under the irradiation of the preset standard light; the endoscopic image imaging method is applicable to the image processing unit, and the endoscopic image imaging method comprises: in the first narrow-band When the light power of the light source changes, an original captured image of the target object captured by the imaging unit under the current light power is obtained; wherein the original captured image is composed of an original first color channel, an original second color channel and an original third color channel; a standard first color channel of the imaging unit corresponding to the standard light power is decomposed in the original first color channel; the standard first color channel, the original second color channel and the original third color channel are fused to obtain a corrected image; brightness information representing brightness and details in the original captured image and color information in the corrected image are respectively extracted, and the brightness information and the color information are fused to obtain an endoscopic imaging image.

[0009] In one embodiment, decomposing the standard first color channel of the imaging unit corresponding to the standard light power in the original first color channel includes: decomposing at least one original first color sub-channel of the imaging unit corresponding to the at least one first narrow-band light source at the current light power in the original first color channel, wherein each original first color sub-channel corresponds to a first narrow-band light source; decomposing the standard first color sub-channels of the imaging unit corresponding to the first narrow-band light source at the standard light power in the original first color sub-channels; and fusing the standard first color sub-channels to obtain the standard first color channel.

[0010] In one embodiment, the first wavelength band includes a narrowband wavelength of 415±10nm and / or a narrowband wavelength of 460±10nm, and correspondingly, the first narrowband light source includes a narrowband blue-violet light source with a central wavelength of 415±10nm and / or a narrowband blue light source with a central wavelength of 460±10nm; the original first color channel includes an original blue channel; the original first color sub-channel includes an original sub-blue channel; the standard first color sub-channel includes a standard sub-blue channel, and the standard first color channel includes a standard blue channel.

[0011] In one embodiment, the decomposing of at least one original sub-first color channel of the imaging unit corresponding to the at least one first narrow-band light source at the current light power in the original first color channel comprises: using a spectral reconstruction method to decompose the original blue channel into the original sub-blue channels of the imaging unit corresponding to the narrow-band blue-violet light source and the narrow-band blue light source at the current light power; or, inputting the original blue channel into a pre-trained original sub-blue channel decomposition model to obtain the original sub-blue channels of the imaging unit corresponding to the narrow-band blue-violet light source and the narrow-band blue light source at the current light power, wherein the original sub-blue channel decomposition model is based on , , The data set is composed of Mapping or The mapping relationship is obtained when The first preset blue channel image data corresponding to each power level combination of the narrow-band blue-violet light source and the narrow-band blue light source is obtained by only enabling the narrow-band blue-violet light source and the narrow-band blue light source at the same time; wherein, The preset blue channel image data corresponding to the i-th narrow-band blue-violet light source power level and the j-th narrow-band blue light source power level; The second preset blue channel data is obtained by enabling the narrow-band blue-violet light source alone and collecting the second preset blue channel data under all adjustable power levels of the narrow-band blue-violet light source; The third preset blue channel data is obtained by enabling the narrow-band blue light source alone and collecting the third preset blue channel data at all adjustable power levels of the narrow-band blue light source.

[0012] In one embodiment, the decomposing the standard first color sub-channels of the imaging unit corresponding to the first narrow-band light source at the standard light power in the original first color sub-channels respectively comprises: using the calculation theory of color constancy perception to decompose the standard sub-blue channels of the imaging unit corresponding to the narrow-band blue-violet light source and the narrow-band blue light source at the standard light power in the original blue sub-channels respectively; or, inputting each of the original blue sub-channels into a standard blue sub-channel decomposition model to obtain a standard blue sub-channel corresponding to the original blue sub-channel; wherein the standard blue sub-channel decomposition model is based on , The data set is trained; The blue channel data under each brightness level is obtained by collecting data at each white light brightness level, keeping other light source levels unchanged, only changing the narrow-band blue-violet light source or the narrow-band blue light source corresponding to the original sub-blue channel, traversing the white light brightness level and all adjustable levels of the narrow-band blue-violet light source or the narrow-band blue light source, and representing a preset standard sub-blue channel image corresponding to the i-th white light brightness level and the j-th level of the narrow-band blue-violet light source or the narrow-band blue light source; the The standard sub-blue channel data corresponding to each white light brightness level is obtained by collecting the standard sub-blue channel data corresponding to each white light brightness level, and the preset sub-standard channel image corresponding to the ith white light brightness level is represented; in the process of training the standard sub-blue channel decomposition model, the corresponding sub-blue channel image under the ith white light brightness level is learned. Mapping The mapping relationship is obtained.

[0013] In one embodiment, the fusing of the standard first color sub-channels to obtain the standard first color channel includes: obtaining preset sub-fusion weights of each of the standard sub-blue channels when fusing; wherein each of the preset sub-fusion weights is determined by obtaining a synthetic blue channel image at a target power corresponding to each of the standard sub-blue channels and a decomposed blue channel image when the narrow-band blue light source and the narrow-band blue-violet light source are separately enabled based on the target power corresponding to each of the standard sub-blue channels, and by fusing the decomposed blue channel images into a blue channel image; and fusing based on the preset sub-fusion weights corresponding to each of the standard sub-blue channels to obtain the standard blue channel.

[0014] In one embodiment, the fusion of the standard first color sub-channels to obtain the standard first color channel further includes: inputting each of the standard sub-blue channels into a standard sub-blue channel fusion model to obtain the standard blue channel; wherein the standard sub-blue channel fusion model is based on , and The data set is trained; It is obtained by enabling the narrow-band blue-violet light source alone, traversing the adjustable gears corresponding to the white light brightness gear where the narrow-band blue-violet light source is located, and collecting the blue channel image data corresponding to each gear; it represents the first standard sub-blue channel image corresponding to the brightness gear of the i-th narrow-band blue-violet light source; the It is obtained by enabling the narrow-band blue light source alone, traversing the adjustable gears corresponding to the white light brightness gear where the narrow-band blue light source is located, and collecting the blue channel image data corresponding to each gear; it represents the second standard sub-blue channel image corresponding to the brightness gear of the i-th narrow-band blue light source; It is obtained by traversing all adjustable gears determined by the narrow-band blue-violet light source and the narrow-band blue light source according to the preset white light brightness gear, and collecting the blue channel data corresponding to the preset white light brightness gear; it represents the synthetic standard blue channel image corresponding to the brightness gear of the i-th group of narrow-band blue-violet light sources and the brightness gear of the narrow-band blue light source; in the process of training the standard sub-blue channel fusion model, learning the mapping relationship when mapping the first standard sub-blue channel image corresponding to the i-th narrow-band blue-violet light source brightness gear and the second standard sub-blue channel image corresponding to the i-th narrow-band blue light source brightness gear to the synthetic standard blue channel image corresponding to the i-th group of narrow-band blue-violet light source brightness gear and the narrow-band blue light source brightness gear.

[0015] In one embodiment, the extracting brightness information representing brightness and details in the original acquired image and color information in the corrected image respectively, and fusing the brightness information and the color information to obtain an endoscopic imaging image, includes: mapping the original acquired image and the corrected image to independent color spaces, and obtaining a first mapping image corresponding to the original acquired image and a second mapping image corresponding to the corrected image respectively; wherein the independent color space includes at least one of a YUV space, a Lab space, and an HSV space; extracting a first channel containing brightness information in the first mapping image; extracting a second channel containing color information and a third channel containing saturation information in the second mapping image; and fusing the first channel, the second channel, and the third channel to obtain the endoscopic imaging image.

[0016] In one embodiment, before acquiring the original captured image of the target object captured by the imaging unit under the current light power, it also includes: acquiring an initial imaging image of the target object and a light power coefficient table corresponding to the target object under all brightness levels of the preset standard light; wherein the initial imaging image is composed of an initial blue channel, an initial red channel, and an initial green channel; the light power coefficient table is used to indicate the power coefficients and power values ​​corresponding to each narrow-band light source constituting the initial blue channel, the initial red channel, and the initial green channel when the target object is imaged under all brightness levels of the preset standard light; based on the initial The red channel and the initial green channel determine the red light power coefficient and red light power value corresponding to the red light source and the green light power coefficient and green light power value corresponding to the green light source; in the light power coefficient table, based on the red light power coefficient, the red light power value, the green light power coefficient and the green light power value, find the standard blue light power value of the blue light source corresponding to the current white light gear; extract the actual blue light power value in the initial imaging image; when the actual blue light power value does not match the standard blue light power value, confirm that the current light power has changed, and use the initial imaging image as the original acquired image.

[0017] According to the second aspect, an embodiment of the present application provides an endoscope, comprising a lighting unit, an imaging unit and an image processing unit, wherein the lighting unit comprises at least one first narrow-band light source of at least one first band corresponding to the blue channel in the RGB image captured by the imaging unit and multiple second light sources of multiple second bands corresponding to the red channel and the green channel, the first narrow-band light source and the second light source can synthesize a preset standard light under the corresponding standard light power, and under the irradiation of the preset standard light, the imaging unit can capture a standard image; the image processing unit comprises a memory for storing a computer program; a processor for implementing the steps of the endoscopic image imaging method as described in any one of the first aspects above when executing the computer program stored in the memory.

[0018] The endoscopic image imaging method provided by the present invention obtains the original captured image of the target object captured by the imaging unit when the optical power of the first narrow-band light source changes; decomposes the standard first color channel of the imaging unit corresponding to the standard optical power in the original first color channel; fuses the standard first color channel, the original second color channel and the original third color channel to obtain a corrected image; respectively extracts the brightness information representing brightness and details in the original captured image and the color information in the corrected image, and fuses the brightness information and the color information to obtain an endoscopic imaging image. The present invention utilizes the different reflection characteristics of the first narrow-band light source at different parts of the human body cavity, and increases or decreases the proportion of the first narrow-band light source for different parts in a targeted manner, thereby causing the original captured image of the target object captured by the imaging unit to be adjusted to the color tone of the preset standard light on the basis of retaining the enhanced detail information brought about by the change in the optical power of the first narrow-band light source when the optical power of the original first color channel changes, so that the adjusted endoscopic imaging image retains the detail enhancement brought about by the change in the proportion of the first narrow-band light source, while maintaining the color expression of the preset standard light imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 4 is a flow chart of an endoscopic imaging method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] According to an embodiment of the present invention, an embodiment of an endoscopic image imaging method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0023] In this embodiment, an endoscopic image imaging method is provided, which can be applied to an endoscopic system; in one embodiment, the endoscopic system includes a lighting unit, an imaging unit, and an image processing unit; the lighting unit includes at least one first narrow-band light source of at least one first band corresponding to a blue channel in an RGB image collected by the imaging unit and multiple second light sources of multiple second narrow-band bands corresponding to a red channel and a green channel, the first narrow-band light source and the second light source can synthesize a preset standard light under the corresponding standard light power, and the imaging unit can collect a standard image under the irradiation of the preset standard light;

[0024] In one embodiment, the lighting unit provides a lighting module for the endoscope system, and the lighting in the target application scenario is only provided by the lighting unit; the imaging unit is used to receive the initial imaging image obtained by reflecting the outgoing light provided by the lighting module after irradiating the target object; wherein the initial imaging image is composed of a blue channel, a red channel and a green channel.

[0025] In one embodiment, the lighting unit is composed of multiple light sources providing different narrowband colors, including at least one first narrowband light source with at least one first band corresponding to the blue channel and multiple second light sources with multiple second narrowband bands corresponding to the red channel and the green channel.

[0026] In this embodiment, the first wavelength band includes a narrowband wavelength band with a central wavelength of 415±10nm and / or a narrowband wavelength band with a central wavelength of 460±10nm, and correspondingly, the first narrowband light source includes a narrowband blue-violet light source with a central wavelength of 415±10nm and / or a narrowband blue light source with a central wavelength of 460±10nm. It should be noted that the first wavelength band can also be other narrowband wavelength bands, which can be determined according to specific application scenarios.

[0027] In the following embodiments, the first wavelength band is described by taking a narrow wavelength band with a central wavelength of 415±10 nm and a narrow wavelength band with a central wavelength of 460±10 nm as an example.

[0028] In one embodiment, the second wavelength band may include a broadband green light source with a central wavelength of 540±40 nm, a broadband amber light source with a central wavelength of 600±30 nm, and a narrowband red light source with a central wavelength of 630±10 nm.

[0029] In this embodiment, the lighting module is a 5LED light source, and the output light output by the lighting module is synthesized by a narrow-band blue light source, a narrow-band blue-violet light source, a broadband green light source, a broadband amber light source and a narrow-band red light source, respectively. The central wavelength range of the narrow-band blue-violet light source is 415±10nm, the central wavelength range of the narrow-band blue light source is 460±10nm, the central wavelength of the broadband green light source is 540±40nm, the central wavelength of the broadband amber light source is 600±30nm, and the central wavelength of the narrow-band red light source is 630±10nm. The blue channel of the initial imaging image received by the imaging unit is synthesized by the narrow-band blue light source and the narrow-band blue-violet light source as an example to illustrate the technical solution of the present invention.

[0030] For example, Figure 1 is a flow chart of an endoscopic imaging method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0031] Step S101, when the optical power of the first narrow-band light source changes, obtain an original captured image of the target object captured by the imaging unit under the current optical power; wherein the original captured image is composed of an original first color channel, an original second color channel and an original third color channel.

[0032] As mentioned above, since different parts of the human body cavity have different absorption and reflection characteristics for blue light and blue-violet light, the ratio of blue light and blue-violet light is increased or decreased specifically for different parts; when the ratio of blue light and blue-violet light changes, the optical power of the original blue channel will change. At this time, it is necessary to adjust the original captured image of the target object captured by the imaging unit to adapt to the white light tone while retaining the enhanced detail information brought about by the change in optical power, so that the adjusted endoscopic imaging image retains the detail enhancement brought about by the change in the ratio of the narrow-band light source, while maintaining the color expression of the white light imaging.

[0033] The original first color channel is described by taking the original blue channel as an example, the original second color channel is described by taking the original red channel as an example, and the original third color channel is described by taking the original green channel as an example. In other embodiments, the first color, the second color, and the third color can be replaced according to different absorption and reflection characteristics of the target object.

[0034] For example, for a lighting module composed of 5 LEDs, the best color rendering index (CRI) and other key indicators of the light source are usually obtained by modulating the optimal ratio of each light source; during the modulation process, the light power parameters of each LED in the 5 LEDs at a certain white light brightness level are determined, and the light power parameter table of each LED corresponding to the lighting module at each white light level can be determined according to the modulation process.

[0035] For the initial image received by the imaging unit, when the ratio of blue light or blue-violet light changes, its actual optical power changes relative to the preset optical power of the narrow-band blue light source and the narrow-band blue-violet light source under the current white light brightness level, while the actual optical power of other light sources providing different colors except the narrow-band blue light source and the narrow-band blue-violet light source remains unchanged relative to the preset optical power of the current white light brightness level; therefore, in one embodiment, the white light brightness level of the current white light illumination can be obtained by matching the first actual optical power of other light sources providing different colors except the narrow-band blue light source and the narrow-band blue-violet light source with the first preset optical power, and then the preset optical power of the narrow-band blue light source and the narrow-band blue-violet light source corresponding to the current white light brightness level can be queried in the optical power parameter table through the white light brightness level. When the actual optical power of the narrow-band blue light source and the narrow-band blue-violet light source does not match the preset optical power, it is confirmed that a change in the optical power of the blue channel has occurred at this time due to the targeted increase or decrease in the ratio of blue light and blue-violet light to different parts.

[0036] In one embodiment, the optical power parameter table may include gear information, optical power values ​​of each LED at each gear, and optical power coefficients.

[0037] Step S102 : decomposing the original first color channel to obtain a standard first color channel of the imaging unit corresponding to the standard light power.

[0038] When the narrow-band blue light source and the narrow-band blue-violet light source change, there is a difference in color between the narrow-band imaging and the white-light imaging; in order to restore the color information of the white-light imaging through the original blue channel, in this embodiment, the standard blue channel of the target object under standard light power is determined based on the original blue channel.

[0039] Since the original blue channel is synthesized by a narrowband blue light source and a narrowband blue-violet light source, in this embodiment, the standard blue channel of the target object relative to the standard light power can be determined based on the changes of the narrowband blue light source and the narrowband blue-violet light source respectively.

[0040] In one embodiment, the original blue channel can be decomposed to obtain original sub-blue channels corresponding to the narrow-band blue light source and the narrow-band blue-violet light source constituting the original blue channel, and the standard sub-blue channel can be obtained by restoring the original sub-blue channels, and the standard sub-blue channels can be further fused to obtain the standard blue channel.

[0041] Among them, as a possible implementation method, the decomposition of the original blue channel can be achieved through deep learning or spectral reconstruction.

[0042] As a possible implementation method, the original sub-blue channel can be restored by deep learning or image processing based on Retinex theory to obtain a standard sub-blue channel.

[0043] As a possible implementation method, the standard sub-blue channels can be fused through deep learning or proportional fusion to obtain a standard blue channel.

[0044] Step S103, fusing the standard first color channel, the original second color channel and the original third color channel to obtain a corrected image.

[0045] In this embodiment, narrow-band imaging is achieved by adjusting the narrow-band blue light source and the narrow-band blue-violet light source corresponding to the blue channel, while the optical power and ratio of the other light sources corresponding to the original red channel and the original green channel are usually not changed during narrow-band imaging, that is, the other light sources corresponding to the original red channel and the original green channel still maintain the optical power and ratio of narrow-band imaging during white light imaging; for the standard blue channel obtained from the original blue channel, it is combined with the original red channel and the original green channel to obtain a color-corrected image of the target object at a given brightness level of white light with standard optical power; wherein the color-corrected image includes color information that can characterize the target object at a given brightness level of white light during white light imaging, and the color information is used to characterize the white light hue of the target object during white light imaging.

[0046] Step S104 , respectively extracting brightness information representing brightness and details in the original acquired image and color information in the corrected image, and fusing the brightness information and the color information to obtain an endoscopic imaging image.

[0047] In order to ensure that the final endoscopic imaging image includes both detail enhancement brought by narrow-band imaging and color correction brought by white-light imaging, in this embodiment, the brightness information in the original acquired image and the color information in the color-corrected image are fused to obtain the endoscopic imaging image.

[0048] In one embodiment, the original captured image and the color corrected image can be converted into independent color spaces respectively, and the brightness information in the channel representing the brightness and the color information in the channel representing the color in the color space can be further extracted, and finally the brightness information and the color information can be fused to obtain the endoscopic imaging image.

[0049] The endoscopic image imaging method provided by the present invention obtains the original captured image of the target object captured by the imaging unit when the light power of the original blue channel changes; wherein the original captured image is composed of the original blue channel, the original red channel and the original green channel; based on the original blue channel, the standard blue channel of the target object is determined relative to the standard light power; the standard blue channel, the original red channel and the original green channel are fused to obtain a color-corrected image of the target object under the standard light power; the brightness information in the original captured image and the color information in the color-corrected image are fused to obtain an endoscopic imaging image. When the light power of the original blue channel changes, the present invention restores the corrected image of the endoscope under a predetermined brightness level of white light based on the original blue channel, further fuses the original captured image containing detail enhancement and the color-corrected image containing color to obtain the endoscopic imaging image, and retains the enhanced details while restoring the white light tone.

[0050] The technical solutions described in the above embodiments are implemented based on narrow-band imaging when changing the narrow-band blue light source and the narrow-band blue-violet light source; it should be understood that for example, for solutions of changing the blue light source or the blue-violet light source alone, or for changing other single light sources or all other light sources contained in the lighting module alone or together to achieve other types of imaging solutions, the original channels can be restored to obtain their corresponding standard channels by changing the blue channel, the red channel and / or the green channel brought about by the change of the light source, and the information contained in the standard channel is further fused with the information contained in the original channel to restore the image corresponding to the white light imaging. Finally, by fusing the original image containing the additional beneficial effects brought about by the light source change and the color-corrected image that can restore the endoscope at a given brightness level of white light, it is achieved that the additional beneficial effects brought about by the light source change are retained while restoring the white light hue.

[0051] As an exemplary embodiment, when the optical power of the original blue channel changes, determining the standard blue channel of the target object relative to the standard optical power based on the original blue channel includes: decomposing the original blue channel to obtain original sub-blue channels constituting the original blue channel; wherein the original sub-blue channels are respectively provided by a narrow-band blue-violet light source with a central wavelength of 415±10nm and a narrow-band blue light source with a central wavelength of 460±10nm; decomposing each of the original sub-blue channels to obtain a standard sub-blue channel corresponding to each of the original sub-blue channels; and fusing each of the standard sub-blue channels to obtain the standard blue channel.

[0052] Since the original blue channel is synthesized by a narrow-band blue light source and a narrow-band blue-violet light source, in this embodiment, the original blue channel is decomposed to obtain original sub-blue channels constituting the original blue channel, and each of the original sub-blue channels is further decomposed to obtain a standard sub-blue channel corresponding to each of the original sub-blue channels, and finally the standard blue channel is obtained by fusing each of the standard sub-blue channels.

[0053] As described above, the decomposition of the original blue channel can be achieved by deep learning; based on this, as an exemplary embodiment, the decomposition of at least one original sub-first color channel of the imaging unit corresponding to the at least one first narrow-band light source at the current light power in the original first color channel includes: inputting the original blue channel into a pre-trained original sub-blue channel decomposition model to obtain the original sub-blue channels of the imaging unit corresponding to the narrow-band blue-violet light source and the narrow-band blue light source at the current light power, and the original sub-blue channel decomposition model is based on , and The data set is composed of Mapping or The mapping relationship is obtained when The first preset blue channel image data corresponding to each gear combination of the narrow-band blue-violet light source and the narrow-band blue light source is obtained by only enabling the narrow-band blue-violet light source and the narrow-band blue light source at the same time; wherein, The preset blue channel image data corresponding to the i-th narrow-band blue-violet light source brightness level and the j-th narrow-band blue light source brightness level; The second preset blue channel data is obtained by enabling the narrow-band blue-violet light source alone and collecting the second preset blue channel data under all adjustable gears of the narrow-band blue-violet light source; The third preset blue channel data is obtained by enabling the narrow-band blue light source alone and collecting the third preset blue channel data under all adjustable gears of the narrow-band blue light source.

[0054] In this embodiment, the original sub-blue channel decomposition model is used to decompose the original blue channel to obtain the original sub-blue channels constituting the original blue channel. The original sub-blue channel decomposition model can be a deep learning model, specifically, a neural network model (NN). The deep learning model consists of , and The constructed data set is trained.

[0055] Exemplarily, the technical solution of this embodiment is described by taking the training of obtaining the first original sub-blue channel corresponding to the narrow-band blue-violet light source by decomposing the original blue channel as an example:

[0056] In the target scene, only the narrow-band blue-violet light source and the narrow-band blue light source are enabled, and the adjustable gears of the two light sources are traversed respectively. The corresponding blue channel image data under each gear combination is collected as the training data, which is recorded as , represents the synthetic blue channel image corresponding to the brightness level of the i-th narrow-band blue-violet light source and the brightness level of the j-th narrow-band blue light source.

[0057] Enable the narrow-band blue-violet light source separately, traverse the adjustable gears of the light source, and collect the corresponding blue channel image data under each gear as the ground truth of the training data, denoted as , represents the blue channel image corresponding to the brightness level of the i-th narrow-band blue-violet light source.

[0058] Further, and As image pairs, they are organized into training data sets; using deep learning methods, the training data is input into the pre-built NN for model training. During the model training process, the learning Mapping , until the model converges.

[0059] It should be understood that by Replace with The implementation method realizes the training of the original sub-blue channel decomposition model of the second original sub-blue channel corresponding to the narrow-band blue light source obtained by decomposing the original blue channel, which will not be repeated here.

[0060] As described above, the decomposition of the original blue channel can also be achieved by spectral reconstruction; based on this, as an exemplary embodiment, the decomposition of the original blue channel to obtain the original sub-blue channels constituting the original blue channel also includes: obtaining the single-component estimation matrix corresponding to each of the original sub-blue channels; wherein the single-component estimation matrix is ​​determined by at least one of the R matrix method, the principal component analysis method, the Wiener estimation method, and the comparative measurement method; and the original acquired image is multiplied by each of the single-component estimation matrices to obtain each of the original sub-blue channels.

[0061] In this embodiment, for the original captured image, it is assumed that is the single-component estimation matrix related to the system spectral response characteristics of the narrow-band blue-violet light source, , is the single-component estimation matrix related to the system spectral response characteristics of the narrow-band blue light source, , , Determined by at least one of an R matrix method, a principal component analysis method, a Wiener estimation method, and a comparative measurement method, the original blue channel can be decomposed by the following formula to obtain original sub-blue channels constituting the original blue channel:

[0062]

[0063]

[0064] In the formula, Represents the first original sub-blue channel corresponding to the narrow-band blue-violet light source, Represents the second original blue sub-channel corresponding to the narrow-band blue light source, represents the single-component estimation matrix associated with the system spectral response characteristics of a narrow-band blue-violet light source, is a single component estimate related to the system spectral response characteristics of a narrow-band blue light source, R represents the original red channel, G represents the original green channel, Represents the original blue channel.

[0065] Since both the deep learning model and the method based on spectral reconstruction have their own errors when performing blue channel decomposition, in one embodiment, the decomposition of the original blue channel can be achieved simultaneously by the method based on deep learning and the method based on spectral reconstruction; specifically, after obtaining the first original blue channel decomposition result based on the deep learning method and the second original blue channel decomposition result based on the spectral reconstruction method, the first error of the original sub-blue channel decomposition model when performing blue channel decomposition and the second error based on the spectral reconstruction method are obtained respectively, and the first error and the second error are further combined to fuse the first original blue channel decomposition result and the second original blue channel decomposition result to obtain a corrected original channel decomposition result, so as to achieve the decomposition of the original blue channel; wherein, as a possible implementation method, their respective fusion weights are inversely proportional to their errors.

[0066] As described above, the original sub-blue channel can be restored by a deep learning method to obtain a standard sub-blue channel; based on this, as an exemplary embodiment, each of the original sub-blue channels is decomposed to obtain a standard sub-blue channel corresponding to each of the original sub-blue channels, including: inputting each of the original sub-blue channels into a standard sub-blue channel decomposition model to obtain a standard sub-blue channel corresponding to the original sub-blue channel; wherein the standard sub-blue channel decomposition model is based on , The data set is trained; The blue channel data under each brightness level is obtained by collecting data at each white light brightness level, keeping other light source levels unchanged, only changing the narrow-band blue-violet light source or the narrow-band blue light source corresponding to the original sub-blue channel, traversing the white light brightness level and all adjustable levels of the narrow-band blue-violet light source or the narrow-band blue light source, and representing a preset standard sub-blue channel image corresponding to the i-th white light brightness level and the j-th level of the narrow-band blue-violet light source or the narrow-band blue light source; the The standard sub-blue channel data corresponding to each white light brightness level is obtained by collecting the standard sub-blue channel data corresponding to each white light brightness level, and the preset sub-standard channel image corresponding to the ith white light brightness level is represented; in the process of training the standard sub-blue channel decomposition model, the corresponding sub-blue channel image under the ith white light brightness level is learned. Mapping The mapping relationship is obtained.

[0067] In this embodiment, the standard sub-blue channel decomposition model is used to decompose the original sub-blue channel to obtain the standard sub-blue channel corresponding to the original sub-blue channel. The standard sub-blue channel decomposition model can be a deep learning model, specifically, a neural network model (NN). The deep learning model consists of , The constructed data set is trained.

[0068] Exemplarily, the technical solution of this embodiment is described by taking the training of a standard sub-blue channel decomposition model for obtaining a first standard sub-blue channel by decomposing the first original blue sub-channel corresponding to a narrow-band blue-violet light source as an example:

[0069] In the target scene, the images corresponding to each white light brightness level are collected and the blue channel image data is extracted as the ground truth of the training data, which is recorded as , represents the standard blue channel target image corresponding to the i-th white light brightness level.

[0070] Keep the intensity of other light sources unchanged, only change the intensity of the narrow-band blue-violet light source corresponding to the current brightness level, traverse the adjustable levels of the light source, and obtain a set of training data sets for the light source at the current brightness level, recorded as Represents the blue channel image corresponding to the j-th level of the narrow-band blue-violet light source corresponding to the i-th white light brightness level.

[0071] Will and As image pairs, they are organized into training data sets. Using deep learning methods, the training data is input into the pre-built NN for model training. During the model training process, the learning Mapping , until the model converges to obtain the standard sub-blue channel decomposition model.

[0072] It should be understood that by keeping the intensity of other light sources unchanged and only changing the narrow-band blue light intensity corresponding to the current brightness level, the adjustable levels of the light source can be traversed to obtain a set of training data sets for the light source at the current brightness level, and further training the model to obtain a standard sub-blue channel decomposition model; the standard sub-blue channel decomposition model is used to obtain a second standard sub-blue channel by decomposing the second original blue sub-channel corresponding to the narrow-band blue light source.

[0073] As described above, the restoration of the original sub-blue channel can also be achieved by image processing based on the Retinex theory; based on this, as an exemplary embodiment, the decomposition of each of the original sub-blue channels to obtain a standard sub-blue channel corresponding to each of the original sub-blue channels includes:

[0074] Decomposing the original collected image into an original illumination image and an original reflection image;

[0075] Acquire a nonlinear mapping function corresponding to each of the original blue sub-channels; wherein the nonlinear mapping function is used to represent the corresponding relationship between the illumination component of the light source corresponding to the original blue sub-channel and the light source intensity;

[0076] The standard sub-blue channel is calculated using the following formula:

[0077]

[0078] in, represents the original blue channel, f represents the nonlinear mapping function of the original sub-blue channel, represents the original reflection image, Indicates the standard optical power of the light source corresponding to the original blue sub-channel at the current white light brightness level.

[0079] Exemplarily, taking the training of decomposing the first original blue sub-channel corresponding to the narrow-band blue-violet light source to obtain the first standard sub-blue channel as an example, the technical solution of this embodiment is described:

[0080] Based on the Retinex theory, the image I(x,y) is decomposed into an illumination image and a reflection image, which can be expressed as

[0081] I(x, y) =R(x, y)×L(x, y)

[0082] Where I(x,y) represents the image signal being observed or received by the camera, L(x,y) represents the illumination component of the ambient light, and R(x,y) represents the reflection component of the target object carrying the image detail information.

[0083] set up is the first original blue subchannel corresponding to the narrow-band blue-violet light source, is the reflection diagram corresponding to the narrow-band blue-violet light source, is the illumination component of the narrow-band blue-violet light source. Since the reflection component is an inherent property of the object and has nothing to do with lighting and equipment, ;in, It is the first standard sub-blue channel corresponding to the narrow-band blue-violet light source.

[0084] For example, suppose is the current optical power of the narrow-band blue-violet light source, is the standard optical power corresponding to the narrow-band blue-violet light source at the current white light brightness level. The illumination component has a positive correlation with the light source intensity. Let f() be the nonlinear mapping function corresponding to the narrow-band blue-violet light source. The corresponding brightness component when mapped to a device-independent space such as XYZ space can be expressed as:

[0085]

[0086] but

[0087] in, represents the original blue channel, f represents the nonlinear mapping function of the original sub-blue channel, represents the original reflection image, Indicates the standard optical power of the light source corresponding to the original blue sub-channel at the current white light brightness level.

[0088] It should be understood that the second standard blue sub-channel can be further decomposed by the second original blue sub-channel corresponding to the narrow-band blue light source through the above method, which will not be described in detail here.

[0089] Since the deep learning model and the image processing method based on Retinex theory both have their own errors when restoring the original sub-blue channel, in one embodiment, the decomposition of the original blue channel can be achieved by both the deep learning method and the Retinex theory-based image processing method; specifically, after obtaining the first standard blue channel decomposition result based on the deep learning method and the second standard blue channel decomposition result based on the Retinex theory image processing method, the first error of the standard sub-blue channel decomposition model when performing the standard blue channel decomposition and the second error of the Retinex theory-based image processing method are respectively obtained, and the first error and the second error are further combined to fuse the first standard blue channel decomposition result and the second standard blue channel decomposition result to obtain a modified standard channel decomposition result; wherein, as a possible implementation method, their respective fusion weights are inversely proportional to their errors.

[0090] As described above, the fusion of standard sub-blue channels can be achieved by proportional fusion to obtain a standard blue channel; based on this, as an exemplary embodiment, the standard sub-blue channels are fused to obtain the standard blue channel, including: obtaining preset sub-fusion weights of each standard sub-blue channel when fused; wherein each preset sub-fusion weight is determined by obtaining a synthetic blue channel image at a target power corresponding to each standard sub-blue channel and a decomposed blue channel image when the narrow-band blue light source and the narrow-band blue-violet light source are separately enabled based on the target power corresponding to each standard sub-blue channel, and by fusing the decomposed blue channel images into a blue channel image; fusion is performed based on the preset sub-fusion weights corresponding to each standard sub-blue channel to obtain the standard blue channel.

[0091] Specifically, since the power values ​​of each light source corresponding to each white light brightness level are determined, in the target scene, a synthetic blue channel image can be obtained when the narrow-band blue-violet light source and the narrow-band blue light source are enabled at the target power at the same time. And the blue channel image based on the target power of the narrow-band blue-violet light source and the narrow-band blue light source when each is enabled separately and , can be obtained through , and The fusion ratio coefficient is fitted by and to obtain the first preset sub-fusion weight of the first standard sub-blue channel and the second standard sub-blue channel when fused, and β represents the second preset sub-fusion weight of the second standard sub-blue channel when fused.

[0092] Furthermore, the fusion of the standard sub-blue channel can be achieved by the following formula:

[0093]

[0094] In the formula, represents the standard blue channel, represents the first standard blue sub-channel, α represents the first preset sub-fusion weight of the first standard blue sub-channel during fusion, represents the second standard sub-blue channel, and β represents the second preset sub-fusion weight of the second standard sub-blue channel during fusion.

[0095] As described above, the fusion of the standard sub-blue channels can be achieved by deep learning to obtain the standard blue channel; based on this, as an exemplary embodiment, the fusion of the standard sub-blue channels to obtain the standard blue channel also includes:

[0096] Each of the standard sub-blue channels is input into a standard sub-blue channel fusion model to obtain the standard blue channel; wherein the standard sub-blue channel fusion model is based on , and The constructed data set is trained;

[0097] Said It is obtained by enabling the narrow-band blue-violet light source alone, traversing the adjustable gears corresponding to the white light brightness gear where the narrow-band blue-violet light source is located, and collecting the blue channel image data corresponding to each gear; it represents the first standard sub-blue channel image corresponding to the i-th narrow-band blue-violet light source brightness gear;

[0098] Said It is obtained by enabling the narrow-band blue light source alone, traversing the adjustable gears corresponding to the white light brightness gear where the narrow-band blue light source is located, and collecting the blue channel image data corresponding to each gear; it represents the second standard sub-blue channel image corresponding to the brightness gear of the i-th narrow-band blue light source;

[0099] Said It is obtained by traversing all adjustable gears determined by the narrow-band blue-violet light source and the narrow-band blue light source according to the preset white light brightness gear, and collecting the blue channel data corresponding to the preset white light brightness gear; it represents the synthetic standard blue channel image corresponding to the brightness gear of the i-th group of narrow-band blue-violet light sources and the brightness gear of the narrow-band blue light source; in the process of training the standard sub-blue channel fusion model, learning the mapping relationship when mapping the first standard sub-blue channel image corresponding to the i-th narrow-band blue-violet light source brightness gear and the second standard sub-blue channel image corresponding to the i-th narrow-band blue light source brightness gear to the synthetic standard blue channel image corresponding to the i-th group of narrow-band blue-violet light source brightness gear and the narrow-band blue light source brightness gear.

[0100] As an exemplary embodiment, the respectively extracting brightness information representing brightness and details in the original acquired image and color information in the corrected image, and fusing the brightness information and the color information to obtain an endoscopic imaging image, includes: mapping the original acquired image and the standard image to independent color spaces, and respectively obtaining a first mapping image corresponding to the original acquired image and a second mapping image corresponding to the corrected image; wherein the independent color space includes at least one of a YUV space, a Lab space, and an HSV space; extracting a first channel containing brightness information in the first mapping image; extracting a second channel containing color information and a third channel containing saturation information in the second mapping image; and fusing the first channel, the second channel, and the third channel to obtain the endoscopic imaging image.

[0101] In this embodiment, by extracting the first channel containing brightness information in the first mapping image, extracting the second channel containing color information and the third channel containing saturation information in the second mapping image, and then fusing the multiple channels, it is possible to fuse the original acquired image containing detail enhancement and the corrected image containing color to obtain an endoscopic imaging image, which can restore the white light tone while retaining the enhanced details.

[0102] In one embodiment, after extracting the second channel containing color information and the third channel containing saturation information in the second mapping image, before fusing the first channel, the second channel and the third channel, it also includes: using a stretching algorithm to process the third channel containing saturation information to obtain an enhanced third channel with improved contrast; fusing the first channel, the second channel and the enhanced third channel to obtain a first contrast-enhanced endoscopic imaging image.

[0103] As a possible implementation, the third channel is subjected to guided filtering processing using the first channel as a guide map to obtain an enhanced third channel.

[0104] In one embodiment, before fusing the first channel, the second channel and the third channel, it also includes: using a stretching algorithm to obtain an enhanced first channel, an enhanced second channel and an enhanced third channel with improved contrast; fusing the enhanced first channel, the enhanced second channel and the enhanced third channel to obtain a second contrast-enhanced endoscopic imaging image.

[0105] In order to determine whether the current endoscope system performs narrow-band imaging, as an exemplary embodiment, before acquiring the original captured image of the target object captured by the imaging unit, it also includes: acquiring an initial imaging image of the target object and an optical power coefficient table corresponding to the target object at each white light brightness level; wherein the initial imaging image is composed of an initial blue channel, an initial red channel, and an initial green channel; the optical power coefficient table is used to indicate the power coefficients and power values ​​corresponding to the light sources constituting the initial blue channel, the initial red channel, and the initial green channel when the target object is imaged at each white light brightness level; based on the initial The red channel and the initial green channel determine the red light power coefficient and red light power value corresponding to the red light source and the green light power coefficient and green light power value corresponding to the green light source; in the light power coefficient table, based on the red light power coefficient, the red light power value, the green light power coefficient and the green light power value, find the standard blue light power value of the blue light source corresponding to the current white light gear; extract the actual blue light power value in the initial imaging image; when the actual blue light power value does not match the standard blue light power value, confirm that the light power of the original blue channel has changed, and use the initial imaging image as the original acquired image.

[0106] When the endoscope system performs narrow-band imaging, only the power value and power coefficient of the narrow-band blue-violet light source and the narrow-band blue light source change under the current white light gear, and the power values ​​and power coefficients of other light sources remain unchanged; therefore, in this embodiment, the red light power coefficient and red light power value corresponding to the red light source and the green light power coefficient and green light power value corresponding to the green light source are determined based on the initial red channel and the initial green channel, and then the current white light gear can be determined according to the power coefficient and power value corresponding to the initial red channel and the initial green channel.

[0107] Furthermore, the standard blue light power value of the blue light source corresponding to the current white light level is searched in the optical power coefficient table according to the white light level.

[0108] When the actual blue light power value in the initial imaging image does not match the standard light power value, it is confirmed that the light power of the original blue channel has changed, and the initial acquired image needs to be processed at this time; based on this, the initial acquired image is used as the original acquired image, and the original acquired image is processed by the endoscopic image imaging method in the above embodiment to obtain an endoscopic imaging image, so that the processed image restores the white light tone while retaining the enhanced details.

[0109] According to the second aspect, an embodiment of the present application provides an endoscope, comprising a lighting unit, an imaging unit and an image processing unit, wherein the lighting unit comprises at least one first narrow-band light source of at least one first band corresponding to the blue channel in the RGB image captured by the imaging unit and multiple second light sources of multiple second bands corresponding to the red channel and the green channel, the first narrow-band light source and the second light source can synthesize preset standard light under the corresponding standard light power, and under the irradiation of the preset standard light, the imaging unit can capture a standard image; the image processing unit comprises a memory for computer programs; a processor for implementing the steps of the endoscopic image imaging method as described in any one of the first aspects above when executing the computer program stored in the memory.

[0110] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0111] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which may be personal computers, servers or network devices, etc.) to execute all or part of the steps of the method in the above embodiments.

[0112] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0113] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in this embodiment.

[0114] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0115] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0116] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An endoscopic imaging method, characterized in that: The endoscope comprises an illumination unit, an imaging unit and an image processing unit, wherein the illumination unit comprises at least one first narrow-band light source of at least one first band corresponding to a blue channel in an RGB image acquired by the imaging unit and a plurality of second light sources of a plurality of second bands corresponding to a red channel and a green channel, the first narrow-band light source and the second light source being able to synthesize a preset standard light under corresponding standard light power, and the imaging unit being able to acquire a standard image under the irradiation of the preset standard light; The endoscopic image imaging method is applicable to the image processing unit, and the endoscopic image imaging method comprises: When the optical power of the first narrow-band light source changes, an original captured image of the target object captured by the imaging unit under the current optical power is acquired; wherein the original captured image is composed of an original first color channel, an original second color channel, and an original third color channel; Decomposing the original first color channel into a standard first color channel of the imaging unit corresponding to the standard light power; fusing the standard first color channel, the original second color channel and the original third color channel to obtain a corrected image; Respectively extracting brightness information representing brightness and details in the original acquired image and color information in the corrected image, and fusing the brightness information and the color information to obtain an endoscopic imaging image; The step of decomposing the original first color channel into a standard first color channel of the imaging unit corresponding to the standard light power comprises: Decomposing at least one original first color sub-channel of the imaging unit corresponding to the at least one first narrow-band light source at the current light power in the original first color channel, wherein each original first color sub-channel corresponds to one first narrow-band light source; Decomposing the standard first color sub-channels of the imaging unit corresponding to the first narrow-band light source at the standard light power in the original first color sub-channels respectively; The standard first color sub-channels are fused to obtain the standard first color channel.

2. The endoscopic imaging method according to claim 1, wherein: The first wavelength band includes a narrowband wavelength of 415±10nm and / or a narrowband wavelength of 460±10nm, and correspondingly, the first narrowband light source includes a narrowband blue-violet light source with a central wavelength of 415±10nm and / or a narrowband blue light source with a central wavelength of 460±10nm; The original first color channel includes an original blue channel; the original first color sub-channel includes an original sub-blue channel; the standard first color sub-channel includes a standard sub-blue channel, and the standard first color channel includes a standard blue channel.

3. The endoscopic imaging method according to claim 2, characterized in that: The step of decomposing at least one original sub-first color channel of the imaging unit corresponding to the at least one first narrow-band light source at the current light power from the original first color channel comprises: Decomposing the original blue channel into the narrow-band blue-violet light source and the original sub-blue channel of the imaging unit corresponding to the narrow-band blue light source at the current light power by using a spectral reconstruction method; or, The original blue channel is input into a pre-trained original sub-blue channel decomposition model to obtain the original sub-blue channels of the imaging unit corresponding to the narrow-band blue-violet light source and the narrow-band blue light source at the current light power, The original sub-blue channel decomposition model is , and The data set is composed of Mapping or The mapping relationship is obtained when Said The first preset blue channel image data corresponding to each power level combination of the narrow-band blue-violet light source and the narrow-band blue light source is obtained by only enabling the narrow-band blue-violet light source and the narrow-band blue light source at the same time; wherein, Representing the preset blue channel image data corresponding to the i-th narrow-band blue-violet light source power level and the j-th narrow-band blue light source power level; Said The second preset blue channel data is obtained by enabling the narrow-band blue-violet light source alone and collecting the second preset blue channel data at all adjustable power levels of the narrow-band blue-violet light source; Said The third preset blue channel data is obtained by enabling the narrow-band blue light source alone and collecting the third preset blue channel data at all adjustable power levels of the narrow-band blue light source.

4. The endoscopic imaging method according to claim 2, wherein: The step of respectively decomposing the standard first color sub-channels of the imaging unit corresponding to the first narrow-band light source at the standard light power from the original first color sub-channels comprises: Decomposing the original sub-blue channel using the computational theory of color constancy perception into standard sub-blue channels of the imaging unit corresponding to the narrow-band blue-violet light source and the narrow-band blue light source at standard light power; or, Each of the original sub-blue channels is input into the standard sub-blue channel decomposition model to obtain a standard sub-blue channel corresponding to the original sub-blue channel; wherein the standard sub-blue channel decomposition model is based on , The constructed data set is trained; Said The blue channel data under each brightness level is obtained by collecting data under each white light brightness level, keeping other light source levels unchanged, only changing the narrow-band blue-violet light source or the narrow-band blue light source corresponding to the original sub-blue channel, traversing the white light brightness level and all adjustable levels of the narrow-band blue-violet light source or the narrow-band blue light source, and representing a preset standard sub-blue channel image corresponding to the i-th white light brightness level and the j-th level of the narrow-band blue-violet light source or the narrow-band blue light source; Said It is obtained by collecting the standard sub-blue channel data corresponding to each white light brightness level with unchanged optical power, and represents the preset sub-standard channel image corresponding to the i-th white light brightness level; In the process of training the standard sub-blue channel decomposition model, learn to convert the corresponding white light brightness level under the i-th white light brightness level Mapping The mapping relationship is obtained.

5. The endoscopic imaging method according to claim 2, wherein: The step of fusing the standard first color sub-channel to obtain the standard first color channel includes: Obtaining preset sub-fusion weights of each of the standard sub-blue channels when being fused; wherein each of the preset sub-fusion weights is determined by obtaining a synthetic blue channel image at a target power corresponding to each of the standard sub-blue channels and a decomposed blue channel image when the narrow-band blue light source and the narrow-band blue-violet light source are separately enabled based on the target power corresponding to each of the standard sub-blue channels, and by fusing the decomposed blue channel images into a blue channel image; The standard blue channel is obtained by performing fusion based on the preset sub-fusion weights corresponding to the standard sub-blue channels.

6. The endoscopic imaging method according to claim 2, wherein: The fusing of the standard first color sub-channel to obtain the standard first color channel further includes: Each of the standard sub-blue channels is input into a standard sub-blue channel fusion model to obtain the standard blue channel; wherein the standard sub-blue channel fusion model is based on , and The constructed data set is trained; Said It is obtained by enabling the narrow-band blue-violet light source alone, traversing the adjustable gears corresponding to the white light brightness gear where the narrow-band blue-violet light source is located, and collecting the blue channel image data corresponding to each gear; it represents the first standard sub-blue channel image corresponding to the i-th narrow-band blue-violet light source brightness gear; Said It is obtained by enabling the narrow-band blue light source alone, traversing the adjustable gears corresponding to the white light brightness gear where the narrow-band blue light source is located, and collecting the blue channel image data corresponding to each gear; it represents the second standard sub-blue channel image corresponding to the brightness gear of the i-th narrow-band blue light source; Said It is obtained by traversing all adjustable gears of the narrow-band blue-violet light source and the narrow-band blue light source determined according to the preset white light brightness gear, and collecting the blue channel data corresponding to the preset white light brightness gear; it represents the synthetic standard blue channel image corresponding to the brightness gear of the i-th group of narrow-band blue-violet light source and the brightness gear of the narrow-band blue light source; In the process of training the standard sub-blue channel fusion model, the mapping relationship when learning to map the first standard sub-blue channel image corresponding to the i-th narrow-band blue-violet light source brightness level and the second standard sub-blue channel image corresponding to the i-th narrow-band blue light source brightness level into the synthetic standard blue channel image corresponding to the i-th group of narrow-band blue-violet light source brightness level and narrow-band blue light source brightness level is obtained.

7. The endoscopic imaging method according to claim 1, characterized in that: The step of respectively extracting brightness information representing brightness and details in the original acquired image and color information in the corrected image, and fusing the brightness information and the color information to obtain an endoscopic imaging image includes: Mapping the original collected image and the corrected image to independent color spaces, respectively obtaining a first mapped image corresponding to the original collected image and a second mapped image corresponding to the corrected image; wherein the independent color space includes at least one of a YUV space, a Lab space, and an HSV space; Extracting a first channel containing brightness information in the first mapped image; Extracting a second channel containing color information and a third channel containing saturation information in the second mapped image; The first channel, the second channel and the third channel are fused to obtain the endoscopic imaging image.

8. The endoscopic imaging method according to claim 1, wherein: Before acquiring the original captured image of the target object captured by the imaging unit under the current light power, the method further includes: Acquire an initial imaging image of the target object and an optical power coefficient table corresponding to the target object at all brightness levels of the preset standard light; wherein the initial imaging image is composed of an initial blue channel, an initial red channel, and an initial green channel; the optical power coefficient table is used to indicate the power coefficients and power values ​​corresponding to each narrow-band light source constituting the initial blue channel, the initial red channel, and the initial green channel when the target object is imaged at all brightness levels of the preset standard light; Determine the red light power coefficient and the red light power value corresponding to the red light source and the green light power coefficient and the green light power value corresponding to the green light source based on the initial red channel and the initial green channel; In the optical power coefficient table, based on the red light power coefficient, the red light power value, the green light power coefficient and the green light power value, searching for a standard blue light power value of the blue light source corresponding to the current white light gear; Extracting an actual blue light power value in the initial imaging image; When the actual blue light power value does not match the standard blue light power value, it is confirmed that the current optical power has changed, and the initial imaging image is used as the original collected image.

9. An endoscope, characterized in that: The device comprises an illumination unit, an imaging unit and an image processing unit, wherein the illumination unit comprises at least one first narrow-band light source of at least one first band corresponding to a blue channel in an RGB image collected by the imaging unit and a plurality of second light sources of a plurality of second bands corresponding to a red channel and a green channel, the first narrow-band light source and the second light source can synthesize a preset standard light under corresponding standard light power, and the imaging unit can collect a standard image under the irradiation of the preset standard light; The image processing unit comprises a memory for storing a computer program; A processor, configured to implement the steps of the endoscopic imaging method as described in any one of claims 1 to 8 when executing a computer program stored in the memory.

Images