Image reconstruction method and system, image processing device and endoscope system

Abstract

The invention relates to an image reconstruction method and system, an image processing device and an endoscope system. The method comprises the steps that an original image of a detection object in a preset wave band is acquired; the original image is acquired by an endoscope imaging system; according to color information in the original image and a first spectral parameter of the endoscope imaging system under the preset wave band, determining first reflectivity information of each position on the detection object under the preset wave band; according to the first reflectivity information and the first spectral parameter, generating a first reconstructed image of the detection object in a target wave band; the target wave band comprises at least one target wavelength, and the target wave band is within the range of the preset wave band. By adopting the method, the image reconstruction under any wave band can be realized, the flexibility of wave band selection of the reconstructed image can be improved, and the hardware cost can be saved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an image reconstruction method, system, image processing device and endoscope system. Background Art

[0002] Endoscopic images are images obtained through endoscopy technology. They are generally collected from inside a living organism and can be used as a basis for evaluating lesions and making diagnoses. In actual applications, it is found that under different wavelengths of light, the characteristic details of the same physiological part may be different. For example, under certain specific wavelengths of light, enhanced imaging of vascular tissue can be achieved.

[0003] In order to achieve imaging effects under specific wavelength illumination, it is often necessary to equip a light source device that emits light of a specific wavelength or band. At this time, the development and manufacturing costs of the light source equipment are relatively high, and the wavelength scalability of the emitted light is relatively poor.

[0004] Therefore, in practical applications, a solution is needed to achieve imaging effects under illumination of a specific wavelength. Summary of the invention

[0005] Based on this, it is necessary to provide an image reconstruction method, system, image processing device and endoscope system to address the technical problems that the above method has high cost and poor scalability of the wavelength of the emitted light.

[0006] In a first aspect, the present application provides an image reconstruction method. The method comprises:

[0007] Acquire an original image of the detection object in a preset band; the original image is acquired by an endoscope imaging system;

[0008] Determine first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band;

[0009] A first reconstructed image of the detection object in a target band is generated according to the first reflectivity information and the first spectral parameter; the target band includes at least one target wavelength, and the target band is within the range of the preset band.

[0010] In one of the embodiments, the endoscopic imaging system includes a light source, a sensor, and an optical component;

[0011] The first spectral parameters include the spectral emissivity of the light source, the spectral sensitivity distribution information of the sensor, and the spectral transmittance of the optical component.

[0012] In one embodiment, the endoscopic imaging system further includes a filter; and the first spectral parameter further includes the spectral transmittance of the filter.

[0013] In one embodiment, determining the first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscopic imaging system in the preset band includes:

[0014] Constructing a pseudo-inverse matrix corresponding to the first spectral parameter in the preset band;

[0015] The first reflectivity information of each position on the detection object in the preset band is determined according to the color information in the original image and the pseudo-inverse matrix.

[0016] In one embodiment, determining the first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the pseudo-inverse matrix includes:

[0017] The color information in the original image and the pseudo-inverse matrix are multiplied to obtain first reflectivity information of each position on the detection object in the preset band.

[0018] In one embodiment, generating a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter includes:

[0019] Based on the first reflectivity information, determining second reflectivity information of each position on the detection object in the target band;

[0020] Determining a second spectral parameter of the endoscopic imaging system in the target band;

[0021] Based on the second reflectivity information of each position and the second spectral parameter, a first reconstructed image of the detection object in the target band is generated.

[0022] In one embodiment, generating a first reconstructed image of the detection object in the target band based on the second reflectivity information of each position and the second spectral parameter includes:

[0023] Multiplying the second reflectivity information and the second spectral parameter of each position respectively to obtain color information corresponding to each position;

[0024] According to the mapping relationship that each position corresponds to one pixel point, the first reconstructed image is obtained based on the color information corresponding to each position.

[0025] In one embodiment, the target wavelength band includes at least two target wavelengths; generating a first reconstructed image of the detection object in the target wavelength band according to the first reflectivity information and the first spectral parameter includes:

[0026] generating a second reconstructed image of the detection object at each of the target wavelengths according to the first reflectivity information and the first spectral parameter;

[0027] The second reconstructed images of all the target wavelengths are synthesized to obtain the first reconstructed image.

[0028] In one embodiment, generating a second reconstructed image of the detection object at each of the target wavelengths according to the first reflectivity information and the first spectral parameter includes:

[0029] For any one to-be-processed wavelength among the target wavelengths, determining third reflectivity information of each position on the detection object at the to-be-processed wavelength based on the first reflectivity information;

[0030] determining a third spectral parameter of the endoscopic imaging system at the wavelength to be processed;

[0031] Based on the third reflectivity information of each position and the third spectral parameter, a second reconstructed image of the detection object at the wavelength to be processed is generated.

[0032] In one embodiment, after generating a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter, the method further includes:

[0033] Based on the first reconstructed image, image enhancement processing is performed on the original image to obtain an enhanced image corresponding to the original image.

[0034] In one embodiment, the original image and the first reconstructed image are both images in RGB color space;

[0035] The performing image enhancement processing on the original image based on the first reconstructed image to obtain an enhanced image corresponding to the original image includes:

[0036] Performing color space conversion processing on the first reconstructed image and the original image respectively to obtain a converted reconstructed image and a converted original image; the color space conversion processing is used to convert from RGB color space to YCC color space;

[0037] Generate a fused image based on the color value of the converted reconstructed image in the Y channel and the color values ​​of the converted original image in the three YCC channels;

[0038] The fused image is converted from the YCC color space back to the RGB color space to obtain the enhanced image.

[0039] In one of the embodiments, the target wavelength range is 590nm-620nm;

[0040] After generating a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter, the method further includes:

[0041] According to the first reconstructed image, bleeding points in the detection object are identified.

[0042] In a second aspect, the present application also provides an image reconstruction system. The system comprises:

[0043] An acquisition unit, used for acquiring an original image of the detection object in a preset band; the original image is acquired by an endoscope imaging system;

[0044] A determination unit, configured to determine first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band;

[0045] A reconstruction unit is used to generate a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter; the target band includes at least one target wavelength, and the target band is within the range of the preset band.

[0046] In a third aspect, the present application further provides an image processing device. The image processing device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0047] Acquire an original image of the detection object in a preset band; the original image is acquired by an endoscope imaging system;

[0048] Determine first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band;

[0049] A first reconstructed image of the detection object in a target band is generated according to the first reflectivity information and the first spectral parameter; the target band includes at least one target wavelength, and the target band is within the range of the preset band.

[0050] In a fourth aspect, the present application further provides an endoscope system, which comprises an endoscope, a display, a light source device, and the image processing device as described in the third aspect.

[0051] The endoscope and the light source device constitute an endoscope imaging system, and the endoscope system includes an endoscope imaging system. The endoscope includes a sensor and an optical component. The light source device includes a light source, and the light source device may also include a filter.

[0052] The above-mentioned image reconstruction method, system, image processing device and endoscope system, after obtaining the original image of the detection object collected by the endoscope imaging system in the preset band, determine the first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band; further generate the first reconstructed image of the detection object in the target band within the range of the preset band according to the first reflectivity information and the first spectral parameter. The method is based on the imaging principle of the endoscopic image, and uses the relationship between the image of the detection object and the reflectivity information of the detection object, as well as the first spectral parameter of the endoscope imaging system, to reversely deduce the first reflectivity information of the detection object in the preset band, thereby realizing image reconstruction in any band within the preset band range through the first reflectivity information and the first spectral parameter. This scheme only needs to collect one original image, does not require special optical imaging equipment, and does not require a priori images, and can realize image reconstruction in any band, which can not only improve the flexibility of band selection of reconstructed images, but also save hardware costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1a is a schematic structural diagram of an endoscope imaging system in one embodiment;

[0054] Figure 1b is a schematic structural diagram of an endoscope imaging system in another embodiment;

[0055] Figure 1c is a structural schematic diagram of an endoscope imaging system in yet another embodiment;

[0056] Figure 2 is a schematic diagram of reconstructed images of different wavelengths obtained by performing image reconstruction on an original image in one embodiment;

[0057] Figure 3 A diagram showing an application environment of an image reconstruction method in an embodiment;

[0058] Figure 4 is a schematic flow chart of an image reconstruction method in one embodiment;

[0059] Figure 5 is a schematic diagram of an original image, a first reconstructed image and an enhanced image in one embodiment;

[0060] Figure 6 is a schematic diagram of an original image and a first reconstructed image used for bleeding point detection in one embodiment;

[0061] Figure 7 is a schematic flow chart of an image reconstruction method in another embodiment;

[0062] Figure 8 is a structural block diagram of an image reconstruction system in one embodiment;

[0063] Fig. 9 FIG. 4 is a diagram showing the internal structure of an image processing device in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0066] It should be noted that the endoscopic imaging system includes: a light source and an endoscope (also referred to as a mirror body), and the endoscope includes: a lens, a sensor, and a light guide. Among them, the light source is generally located in a light source device (the light source device and the endoscope are different devices) for emitting light, and the light source can be a monochromatic light source or a composite color light source. And the number of light sources in the light source device can also be one or more, for example, multiple monochromatic light sources can emit light of different colors respectively to obtain a composite color light that is finally emitted. The lens can be a single lens or a group of multiple lenses. The sensor is used for imaging. The light guide can be an optical fiber or other optical lens group for the transmission of light.

[0067] Among them, the lens is generally located on the side of the sensor close to the detection object. On this basis, the specific position distribution of the lens, sensor and light guide part in the endoscope is not limited here, and can be determined according to the actual situation of the endoscope. For example, in one implementation, the lens and sensor can be arranged at the end of the light guide part close to the detection object, that is, arranged between the detection object and the light guide part (such as the insertion part of a soft endoscope and the insertion part of some rigid endoscopes), such as Figure 1a In another implementation, the lens and the sensor may also be disposed at the end of the light guide portion away from the detection object, that is, disposed between the light source and the light guide portion, such as Figure 1b shown.

[0068] In practical applications, depending on actual needs, the light source device may or may not include a filter. The functions of the filter include: 1. For a composite color light source, the filter can filter the composite color light emitted by the composite color light source, thereby obtaining monochromatic light. 2. For both monochromatic and composite color light sources, the filter can filter the wavelengths of the light emitted by the light source, thereby making the light output cleaner and more controllable. For example, Figure 1a and Figure 1b Schematic diagram of an endoscope imaging system without a filter. Figure 1c A schematic diagram of an endoscope imaging system including a filter is shown. The embodiment of the present application does not make too many restrictions on the type of filter, which can be set according to actual needs. For example, in some embodiments, the filter can include a filter. Among them, when the light source device includes a filter, the endoscope imaging system in the embodiment of the present application also includes a filter.

[0069] First, the principles involved in the embodiments of the present application are explained. Since the same material has different reflection characteristics for light of different wavelengths (frequencies), the images produced for light of different wavelengths (frequencies) are different, and the detail information reflected between different images will be different.

[0070] At the same time, when the endoscope imaging system does not include a filter, the light irradiated to the detection object and the light finally reflected to the sensor will be affected by: the spectral emissivity E(λ) of the light source and the spectral transmittance L(λ) of the optical component. When the sensor finally forms an image, it will be affected by the reflected light it receives and the spectral sensitivity S(λ) of the sensor. Among them, the optical component refers to the optical component that exists in the imaging-related optical path in the endoscope and will affect the transmission of light, such as some optical lenses arranged in the optical path. In an embodiment of the present application, the optical component at least includes a lens. Therefore, when the final image is formed, the resulting image is related to the spectral emissivity E(λ) of the light source, the spectral transmittance L(λ) of the optical component and the spectral sensitivity S(λ) of the sensor.

[0071] When a filter is included in the endoscope imaging system, the light irradiated to the detection object and the light finally reflected to the sensor will be affected by: the spectral emissivity E(λ) of the light source, the spectral transmittance L(λ) of the optical component and the spectral transmittance f(λ) of the filter. Therefore, when the final image is formed, the image is related to the spectral emissivity E(λ) of the light source, the spectral transmittance L(λ) of the optical component, the spectral transmittance f(λ) of the filter and the spectral sensitivity S(λ) of the sensor.

[0072] Specifically, when the endoscope imaging system does not include a filter, assuming that the wavelength range of the light emitted by the light source is a specific wavelength band, the image imaging formula can be expressed as the following formula (1).

[0073] v i =∫E i (λ)L i (λ)S(λ)O(λ)dλ,i=R,G,B (1)

[0074] Where i is any wavelength within a specific band, E i (λ) is the spectral emissivity of the light source, L i (λ) is the spectral transmittance of the optical component; when the light guide part is an optical lens group, the optical component refers to the lens and the optical lens group, S(λ) is the spectral sensitivity distribution of the sensor, O(λ) is the material reflectivity, and v i Refers to the image after the integration of the data corresponding to each wavelength in a specific band. i=R means that the wavelength is within the R channel range (i.e. the red range), i=G means that the wavelength is within the G channel range (i.e. the green range), and i=B means that the wavelength is within the B channel range (i.e. the blue range). That is, the total integration of the wavelength corresponding data under the three channels of R, G, and B can be performed to obtain the final image.

[0075] When the endoscope imaging system includes a filter, the filter will affect the wavelength of the light emitted by the light source, thereby affecting the imaging result. Therefore, in order to more accurately characterize the imaging principle, the wavelength range of the light emitted by the light source is assumed to be a specific wavelength, and the above formula (1) can be adjusted to the following formula (2).

[0076] v i =∫E i (λ)f i (λ)L i (λ)S(λ)O(λ)dλ,i=R,G,B (2)

[0077] Among them, f i (λ) is the spectral transmittance of the filter.

[0078] It should be noted that when a specific band does not cover all RGB, the data within the uncovered band range can be regarded as 0, and mathematical operations can be performed normally.

[0079] In the above formulas (1) and (2), E i (λ), f i (λ), L i (λ) and S(λ) can both be regarded as spectral parameters of the endoscope imaging system. Therefore, the above formulas (1) and (2) can be simplified to obtain:

[0080] v i =H i O (3)

[0081] Among them, H i =f i E i L i S, or H i =E i L i S.

[0082] From the above formula (3), it can be seen that when it is necessary to predict the image in a certain band, it is necessary to know the spectral parameter H of the endoscope imaging system: i and the reflectivity O of the material. After the equipment in the endoscope imaging system is determined, H i is known, so the reflectivity O of the material can be estimated by the image of the material collected by the endoscopic imaging system, as shown in the following formula (4).

[0083]

[0084] Because H is not a square matrix, there is no inverse matrix, so the pseudo-inverse matrix can be used instead of the inverse matrix to find an approximate solution.

[0085] Therefore, the reflectivity of a material under any wavelength can be calculated using the following formula (5):

[0086]

[0087] Furthermore, in determining the spectral parameter H i After obtaining the pseudo-inverse matrix and the reflectivity of the material, the parameter values ​​of each RGB channel of the image at a certain wavelength can be estimated to obtain the image corresponding to each channel, as shown in the following formula (6).

[0088]

[0089] in, Represents the value of the pseudo-inverse matrix corresponding to the wavelength.

[0090] Based on formula (6), the image of any wavelength in the specific wavelength band emitted by the light source can be determined. When the light emitted by the light source is white light, the specific wavelength band is the wavelength range covered by the white light, such as Figure 2 As shown, it is a schematic diagram of the wavelengths of 500nm, 450nm and 600nm respectively obtained by reconstructing the original image.

[0091] Based on the above principles, the present application proposes a method for image reconstruction at any wavelength to solve the problem that the existing technology can only achieve image reconstruction of a single band based on hardware equipment and has poor scalability.

[0092] The core concept of this application is: based on the color information of a single original image taken by an endoscopic imaging system with known spectral parameters in a specific band, and the spectral parameters, the reflectivity information of the detection object in the specific band is reversely estimated (the reflectivity information includes the reflectivity of each pixel). On this basis, for any wavelength of light within the sub-band range belonging to a specific band, the color information of each pixel corresponding to the wavelength can be solved according to the reflectivity information and spectral parameters of the detection object, that is, the corresponding reconstructed image at the wavelength is obtained. In this way, the reconstructed images corresponding to each wavelength within the sub-band range belonging to the specific band can be obtained, thereby obtaining the reconstructed image corresponding to the sub-band.

[0093] The image reconstruction method provided in the embodiment of the present application can be applied to Figure 3 In the application environment shown. Among them, the image processing device 302 communicates with the endoscopic imaging system 304 by wired communication or wireless communication. In the application scenario of the present application, the endoscopic imaging system 304 collects the original image of the detection object in the preset band, and the image processing device 302 obtains the original image from the endoscopic imaging system 304, and then determines the first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscopic imaging system in the preset band acquired in advance, and generates the first reconstructed image of the detection object in the target band according to the first reflectivity information and the first spectral parameter, wherein the target band contains at least one target wavelength and the target band is within the range of the preset band.

[0094] In one embodiment, Figure 4 As shown, an image reconstruction method is provided, which is applied to Figure 3 The image processing device 302 in the embodiment is taken as an example to illustrate, and the following steps are included:

[0095] Step S410, obtaining an original image of the detection object in a preset band; the original image is acquired by an endoscope imaging system.

[0096] The detection object refers to an object detected by an endoscope. The embodiments of the present application do not make too many restrictions on the specific situation of the detection object, which can be determined according to the situation of the endoscope detection object in actual application. For example, in some embodiments, the detection object can be some physiological parts inside a living body, such as vascular tissue, intestines, stomach, trachea, abdominal organs (such as liver, pancreas, kidneys, etc.), bladder or surgical incision, etc.

[0097] It should be understood that in actual applications, the light source may have the ability to emit one or more different light bands. In the embodiment of the present application, the preset band refers to the light band actually emitted by the light source of the endoscope imaging system. That is, the preset band belongs to the range of the band that the light source can emit, but it is not necessarily equal to all the bands that the light source can emit.

[0098] It should be understood that the preset band may be a monochromatic light band (i.e., the light source emits light of only one color), such as a red band, a green band, or a blue band, etc. The preset band may also be a composite color band (i.e., the light source emits composite light of multiple colors). For example, as an optional embodiment of the present application, the preset band may be a white light band, or may be a combination band of any multiple color bands among the red band, the green band, and the blue band.

[0099] Specifically, after determining the light source, sensor, light guide, lens and other devices that constitute the endoscope imaging system, the light source can emit light of a preset wavelength band, which is guided to the detection object through the light guide, and a part of the light is reflected by the detection object, and the reflected light can be transmitted to the sensor through the lens. Thus, the image of the detection object is collected by the sensor, and the image is the original image of the detection object in the preset wavelength band. Furthermore, the image processing device can obtain the original image from the sensor for subsequent processing.

[0100] Step S420, determining first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band.

[0101] The color information is specifically an RGB value, and the color information in the original image may include the RGB value of each pixel in the original image.

[0102] Wherein, when the endoscopic imaging system does not include a filter, the first spectral parameter may include the spectral emissivity of the light source in the endoscopic imaging system, the spectral transmittance of the optical component, and the spectral sensitivity distribution information of the sensor. When the endoscopic imaging system includes a filter, the first spectral parameter may include the spectral emissivity of the light source in the endoscopic imaging system, the spectral transmittance of the optical component, the spectral sensitivity distribution information of the sensor, and the spectral transmittance of the filter.

[0103] The preset wavelength band may be understood as a range of wavelengths, which includes multiple wavelengths. Correspondingly, the first spectral parameter of the endoscopic imaging system in the preset wavelength band may include spectral parameters at multiple wavelengths.

[0104] The first reflectivity information of the detection object refers to the measure of the detection object's reflectivity to the incident light, indicating the proportion of light reflected by the detection object surface. The reflectivity of a substance is affected by many factors. For example, in some embodiments, the reflectivity information of the detection object may mainly depend on the wavelength of light emitted by the light source.

[0105] In a specific implementation, once the various devices constituting the endoscopic imaging system are determined, the first spectral parameter of the endoscopic imaging system can be fixed. According to the RGB value of the original image of the detection object in the preset band, combined with the above-mentioned image imaging formula (3), the first reflectivity information of the detection object in the preset band can be inferred.

[0106] More specifically, since the color information in the original image includes the color information of each pixel in the original image, the reflectance information corresponding to each pixel is obtained based on the color information in the original image and the first spectral parameter, and the reflectance information corresponding to each pixel is used as the first reflectance information of each position on the detection object in the preset band.

[0107] Step S430, generating a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter; the target band includes at least one target wavelength, and the target band is within a range of a preset band.

[0108] The target band can be understood as a sub-band of the preset band. For example, the preset band is λ1~λ 20 , the target band can be any band within the preset band range, such as λ2~λ8, λ4~λ5, etc.

[0109] As an optional embodiment of the present application, the target wavelength band may include only one wavelength, for example, the target wavelength is λ2. In this case, the present application embodiment can realize image reconstruction under a single wavelength.

[0110] In a specific implementation, since the target band is a sub-band of the preset band, after determining the first reflectivity information of each position of the detection object in the preset band and the first spectral parameter of the endoscope imaging system in the preset band, the reflectivity information of each position of the detection object in the sub-band of the preset band, that is, the target band, can be determined from the first reflectivity information, recorded as the second reflectivity information; and the spectral parameter of the endoscope imaging system in the target band can be determined from the first spectral parameter, recorded as the second spectral parameter. Based on the second reflectivity information and the second spectral parameter, combined with the above-mentioned image imaging formula (4), the image of the detection object in the target band is reconstructed to obtain a first reconstructed image.

[0111] The above-mentioned image reconstruction method, after obtaining the original image of the detection object collected by the endoscope imaging system in the preset band, determines the first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band; further generates the first reconstructed image of the detection object in the target band within the range of the preset band according to the first reflectivity information and the first spectral parameter. The method is based on the imaging principle of the endoscopic image, and uses the relationship between the image of the detection object and the reflectivity information of the detection object, as well as the first spectral parameter of the endoscope imaging system to reversely deduce the first reflectivity information of the detection object in the preset band, thereby realizing image reconstruction in any band within the preset band range through the first reflectivity information and the first spectral parameter. This scheme only needs to collect one original image, does not require special light imaging equipment, and does not require a priori images, and can realize image reconstruction in any band, which can not only improve the flexibility of band selection of the reconstructed image, but also save hardware costs. At the same time, since the imaging principles of various light sources are the same, only the corresponding imaging system parameters are changed, the method is also suitable for image reconstruction under special light.

[0112] In an exemplary embodiment, in the above step S420, determining the first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band includes:

[0113] Step S421, constructing a pseudo-inverse matrix corresponding to the first spectral parameter in a preset band.

[0114] Specifically, based on the above image imaging formula (4), that is, O = H -1 V, it can be determined that before calculating the reflectivity of the detection object, the first spectral parameter of the endoscope imaging system in the preset band needs to be converted into an inverse matrix.

[0115] It can be understood that since the first spectral parameter of the endoscope imaging system is not a square matrix, there is no inverse matrix. Therefore, the present application uses a pseudo-inverse matrix instead of an inverse matrix to obtain an approximate solution for the reflectivity information.

[0116] Specifically, if the preset band is λ1~λ n , then the first spectral parameter is in the band λ1~λ n The pseudo-inverse matrix can be expressed as:

[0117]

[0118] Among them, each row element in the pseudo-inverse matrix corresponds to the element value of the spectral parameter at a wavelength. For example, the first row element corresponds to the element value of the spectral parameter at wavelength λ1, the second row element corresponds to the element value of the first spectral parameter at wavelength λ2, and so on.

[0119] Step S422, determining first reflectivity information of each position on the detection object in a preset band according to the color information in the original image and the pseudo-inverse matrix.

[0120] Specifically, after the color information and the pseudo-inverse matrix in the original image are determined, the color information and the pseudo-inverse matrix in the original image may be multiplied to obtain the first reflectivity information of each position on the detection object in a preset band.

[0121] More specifically, since the color information in the original image includes the color information of each pixel in the original image, it is necessary to perform multiplication processing on the color information of each pixel and the pseudo inverse matrix.

[0122] For each pixel, if its color information is recorded as Then each position of the detection object is in the band λ1~λ n The first reflectivity information O1 under the illumination of the light source can be expressed as:

[0123]

[0124] In this embodiment, a pseudo-inverse matrix corresponding to a spectral parameter in a preset band is constructed instead of an inverse matrix to achieve the solution of the first reflectivity information of each position on the detection object, so that image reconstruction at any wavelength or band within the preset band can be achieved based on the first reflectivity information.

[0125] It can be understood that the target wavelength band may include only one target wavelength, or may include two or more target wavelengths. The image reconstruction methods in the two cases are respectively described below.

[0126] When the target band includes only one target wavelength, correspondingly, in an exemplary embodiment, in the above step S430, generating a first reconstructed image of the detection object in the target band according to the first reflectivity information and the first spectral parameter can be specifically implemented by the following steps:

[0127] Step S4311, based on the first reflectivity information, determine the second reflectivity information of each position on the detection object at the target wavelength.

[0128] Specifically, assuming that the first reflectivity information of any position of the detection object is as shown in the above formula (7), assuming that the target wavelength is λ4, then the second reflectivity information O2 at this position of the detection object under the target wavelength λ4 can be expressed as:

[0129]

[0130] Step S4321, determining a second spectral parameter of the endoscope imaging system at the target wavelength.

[0131] The content included in the second spectral parameter is the same as that in the first spectral parameter, except that the corresponding wave bands or wavelengths are different.

[0132] Specifically, the preset band is still λ1~λ n , the target wavelength is λ4, for example, the endoscope imaging system is in the preset wavelength band λ1~λ n The first spectral parameter under can be expressed as:

[0133]

[0134] The second spectral parameter of the endoscopic imaging system at the target wavelength λ4 extracted from the first spectral parameter can be expressed as:

[0135] H2=[h R (λ4) h G (λ4) h B (λ4)] (10)

[0136] Step S4331: Generate a first reconstructed image of the detection object at a target wavelength based on the second reflectivity information and the second spectral parameters at each position.

[0137] In a specific implementation, after determining the second reflectivity information of each position on the detection object at the target wavelength and the second spectral parameters of the endoscope imaging system at the target wavelength, the second reflectivity information and the second spectral parameters of each position are multiplied to obtain the color information of each position; according to the mapping relationship in which each position corresponds to a pixel point, the first reconstructed image of the detection object at the target wavelength is obtained based on the color information of each position.

[0138] In this embodiment, taking into account that both the reflectivity and the spectral parameters are affected by the wavelength, when performing image reconstruction, the second reflectivity information and the second spectral parameters corresponding to the target wavelength are extracted from the first reflectivity information and the first spectral parameters according to the target wavelength. Further combined with the image imaging principle, image reconstruction based on the second reflectivity information and the second spectral parameters at any wavelength within the preset band is achieved. Therefore, there is no need to add hardware equipment, and only one original image of the detection object needs to be collected. This not only achieves the flexibility of wavelength selection for reconstructing the image, but also saves the cost of image reconstruction.

[0139] When the target band contains at least two target wavelengths, there are at least two methods for generating the first reconstructed image. One method is to regard the target band as a wavelength range, determine the spectral parameters and reflectivity information under the wavelength range, and finally reconstruct the image under the target band according to the spectral parameters and reflectivity information under the wavelength range as the first reconstructed image. The other method is to first reconstruct the second reconstructed image under each target wavelength separately, and then perform image synthesis on the second reconstructed images under each target wavelength to obtain the first reconstructed image. The following describes the two implementation methods in detail.

[0140] If the first method is used to generate the first reconstructed image at at least two wavelengths, correspondingly, in an exemplary embodiment, in the above step S430, the first reconstructed image of the detection object at the target wavelength band is generated according to the first reflectivity information and the first spectral parameter, which can be specifically implemented by the following steps:

[0141] Step S4312: Determine second reflectivity information of each position on the detection object in the target band based on the first reflectivity information.

[0142] Specifically, assuming that the first reflectivity information of any position of the detection object is as shown in the above formula (7), assuming that the target band is λ4-λ9, then the second reflectivity information O2 of the detection object at any position can be expressed as:

[0143]

[0144] Step S4322, determining a second spectral parameter of the endoscope imaging system in the target band.

[0145] The content included in the second spectral parameter is the same as that in the first spectral parameter, except that the corresponding wavebands are different.

[0146] Specifically, it is assumed that the endoscope imaging system is in the preset wavelength range λ1~λ n The first spectral parameter under the target band is shown in formula (9). When the target band is λ4~λ9, the second spectral parameter of the endoscope imaging system under the target band λ4~λ9 extracted from the first spectral parameter can be expressed as:

[0147]

[0148] Step S4332: Generate a first reconstructed image of the detection object in the target band based on the second reflectivity information and the second spectral parameters at each position.

[0149] In a specific implementation, after determining the second reflectivity information of each position on the detection object in the target band and the second spectral parameters of the endoscope imaging system in the target band, the second reflectivity information and the second spectral parameters of each position are multiplied to obtain the color information of each position; according to the mapping relationship that each position corresponds to a pixel point, the first reconstructed image of the detection object in the target band is obtained based on the color information of each position.

[0150] In this embodiment, taking into account that both reflectivity and spectral parameters are affected by wavelength, when performing image reconstruction, second reflectivity information and second spectral parameters corresponding to the target band are extracted from the first reflectivity information and the first spectral parameters according to the target band, and further combined with the image imaging principle, image reconstruction based on the second reflectivity information and the second spectral parameters in any band within the preset band is achieved, thereby eliminating the need to add hardware equipment and only requiring the acquisition of an original image of the detection object, which not only achieves the flexibility of band selection for reconstructing the image, but also saves the cost of image reconstruction.

[0151] In some other optional embodiments of the present application, when implementing step 430, the target band can also be divided into multiple sub-bands and / or sub-wavelengths, and image reconstruction operations can be performed on these divided sub-bands and / or sub-wavelengths respectively. Among them, the image reconstruction instructions for a single sub-wavelength can refer to the relevant instructions of the embodiment described in the above steps 4311 to 4331, and the image reconstruction instructions for a sub-band can refer to the relevant instructions of the embodiment described in the above steps 4312 to 4332, which are not repeated here. After completing the image reconstruction of the divided sub-bands and / or sub-wavelengths, the obtained reconstructed images are synthesized to obtain the final first reconstructed image.

[0152] On the basis of the above two embodiments, the embodiments of the present application can flexibly meet the needs of practical applications. Users can flexibly choose whether to reconstruct the sub-bands and sub-wavelengths first according to their needs, so as to obtain the reconstructed images of the corresponding sub-bands or sub-wavelengths first. At the same time, it can save some image reconstruction work for sub-bands or sub-wavelengths that do not need to be reconstructed separately, thereby achieving a balance between the actual reconstruction image needs and costs.

[0153] If the second method is used to generate the first reconstructed images at at least two wavelengths, correspondingly, in an exemplary embodiment, in the above step S430, a first reconstructed image of the detection object in the target band is generated according to the first reflectivity information and the first spectral parameter, and also includes: generating a second reconstructed image of the detection object at each target wavelength according to the first reflectivity information and the first spectral parameter; performing image synthesis on the second reconstructed images of all target wavelengths to obtain a first reconstructed image.

[0154] Specifically, a band can be understood as a collection of multiple wavelengths. When the target band contains at least two target wavelengths, the values ​​of each target wavelength contained in the target band are determined, and image reconstruction is performed for each target wavelength to obtain a second reconstructed image at each target wavelength. The second reconstructed images of all target wavelengths are then subjected to image synthesis processing to obtain a first reconstructed image of the detection object at the target band.

[0155] Furthermore, in an exemplary embodiment, a second reconstructed image of the detection object at each target wavelength is generated based on the first reflectivity information and the first spectral parameter, including: for any one of the target wavelengths to be processed, based on the first reflectivity information, determining the third reflectivity information of each position on the detection object at the wavelength to be processed; determining the third spectral parameter of the endoscopic imaging system at the wavelength to be processed; and generating a second reconstructed image of the detection object at the wavelength to be processed based on the third reflectivity information and the third spectral parameter of each position.

[0156] Specifically, the method for generating the second reconstructed image of the detection object at the wavelength to be processed is similar to the method for generating the first reconstructed image of the detection object at the target wavelength. It is necessary to first determine the reflectance information corresponding to the wavelength to be processed from the first reflectance information as the third reflectance information. Similarly, the third spectral parameter of the endoscope imaging system at the wavelength to be processed can be extracted from the first spectral parameter shown in formula (9). Finally, the third reflectance information and the third spectral parameter of each position are multiplied to obtain the color information of each position; according to the mapping relationship that each position corresponds to a pixel point, based on the color information of each position, the first reconstructed image of the detection object at the wavelength to be processed is obtained.

[0157] In this embodiment, by reconstructing images at different wavelengths respectively and then synthesizing the reconstructed images at each wavelength, it is possible to generate a reconstructed image when the target band contains at least two wavelengths, thereby improving the flexibility of image reconstruction.

[0158] In an exemplary embodiment, after the above step S430 generates a first reconstructed image of the detection object in the target band according to the first reflectivity information and the first spectral parameter, it also includes: based on the first reconstructed image, performing image enhancement processing on the original image to obtain an enhanced image corresponding to the original image.

[0159] It is understandable that the characteristic details of the same physiological part may be different under different wavelengths of light. For example, under certain wavelengths of light, enhanced imaging of vascular tissue can be achieved. Therefore, in order to better evaluate and diagnose lesions in endoscopic images, it is often necessary to perform image enhancement processing on the original images collected by the endoscopic imaging system.

[0160] Based on this, on the basis of the image reconstruction method proposed in the above embodiment of the present application, the target band can be determined according to the evaluation requirements of the detection object, and the first reconstructed image of the detection object in the target band can be reconstructed. The first reconstructed image is used to perform image enhancement processing on the original image to obtain an enhanced image corresponding to the original image.

[0161] Specifically, the original image can be enhanced by performing image fusion processing on the first reconstructed image and the original image. For example, the first reconstructed image can be superimposed on the original image by spatial domain conversion to enhance the original image.

[0162] For example, refer to Figure 5 , is a schematic diagram of an original image, a first reconstructed image in a target band, and an enhanced image after the original image is enhanced by the first reconstructed image, as shown in an embodiment. Figure 5 As shown, from left to right are the original image, the first reconstructed image and the enhanced image in the target band of 500-550nm. It can be seen from the figure that after the enhancement of the first reconstructed image, the vascular tissue in area 510 is clearer than the original image.

[0163] In this embodiment, the image enhancement processing is performed on the original image through the first reconstructed image, so as to improve the imaging effect of the original image, so as to more accurately evaluate the actual situation of the detection object in the original image.

[0164] In an exemplary embodiment, both the original image and the first reconstructed image are images in RGB color space; based on the first reconstructed image, the original image is subjected to image enhancement processing to obtain an enhanced image corresponding to the original image, specifically including: performing color space conversion processing on the first reconstructed image and the original image respectively to obtain a converted reconstructed image and a converted original image; the color space conversion processing is used to convert from RGB color space to YCC color space. Based on the color value of the converted reconstructed image in the Y channel and the color value of the converted original image in the three YCC channels, a fused image is generated. The fused image is converted from the YCC color space back to the RGB color space to obtain an enhanced image.

[0165] Specifically, the process of performing image enhancement processing on the original image may include:

[0166] (1) Convert the original image from RGB color space to YCC color space to obtain the converted original image.

[0167] (2) Convert the first reconstructed image from the RGB color space to the YCC color space to obtain a converted reconstructed image.

[0168] (3) The color values ​​of the converted original image and the converted reconstructed image in the Y channel are fused to obtain a fused color value of the Y channel.

[0169] Specifically, the color values ​​of the converted original image and the converted reconstructed image in the Y channel can be weighted fused, that is, Y new =w×Y 原始 +(1-w)×Y 重建 , where Y new Indicates the fused color value of the Y channel, Y 原始 Represents the original image after transformation, Y 重建 It represents the reconstructed image after conversion, w is the weight, which ranges from 0 to 1 and can control the enhancement strength.

[0170] (4) Based on the fused color value of the Y channel and the color values ​​of the two C channels of the converted original image, a fused image is obtained.

[0171] Specifically, the fused color value of the Y channel can be used as the color value of the Y channel in the fused image, and the color values ​​of the two C channels of the converted original image can be used as the color values ​​of the corresponding two C channels in the fused image, thereby obtaining a fused image.

[0172] (5) Convert the fused image from the YCC color space back to the RGB color space to obtain an enhanced image.

[0173] In this embodiment, the original image and the first reconstructed image are first converted from the RGB color space to the YCC color space, and the images are fused in the YCC color space. After the fusion is completed, the fused image is converted from the YCC color space back to the RGB color space. Through this spatial domain conversion method, the first reconstructed image is superimposed on the original image to enhance the original image. This is simple and easy, has high real-time performance, and can retain image details.

[0174] It can be understood that in the medical field, the detection of bleeding points is crucial to locate the specific location and degree of bleeding points, and the accuracy of bleeding point detection is closely related to the clarity of the image. Therefore, in order to improve the detection accuracy of bleeding points, it is necessary to enhance the information of bleeding points. There are two main existing enhancement methods. One is to enhance bleeding points by special light, such as red dichromatic imaging technology (Red Dichromatic Imaging, RDI), but special light has high requirements for hardware design. The other is algorithm-based enhancement, such as the FICE system (Flexible spectral imaging color enhancement, intelligent color enhancement technology). The FICE system assigns specific wavelengths to each R, G, and B component between 400nm and 695nm, with an increment of 5-nm. However, FICE only gives the enhanced image and still cannot remove the influence of blood stains. At the same time, FICE reconstruction requires some prior images as input before the reconstructed image can be estimated, but it is very difficult to obtain clinical images in reality. At the same time, when bleeding point detection is performed by machine learning methods, when the bleeding point is covered by blood stains, it will affect the efficiency of bleeding point detection.

[0175] To address these problems, the image reconstruction method described in the above embodiment of the present application can also be applied. Specifically, in an exemplary embodiment, the target band is 590nm-620nm; after generating a first reconstructed image of the detection object in the target band according to the first reflectivity information and the first spectral parameter in step S430, the method further includes: identifying bleeding points existing in the detection object according to the first reconstructed image.

[0176] In a specific implementation, the target band of 590nm-620nm can be understood as the best band for detecting bleeding points. Therefore, in the application of bleeding point detection, after collecting the original image of the detection object, a first reconstructed image of the detection object in the band of 590nm-620nm can be reconstructed, and bleeding point detection is performed on the first reconstructed image to identify the bleeding points in the detection object. The detection method for detecting bleeding points can be a method based on image segmentation and a supervised learning algorithm, etc., and this application does not make specific restrictions on this.

[0177] refer to Figure 6 , is a schematic diagram of comparison between an original image and a first reconstructed image in a target band of 590nm-620nm in one embodiment, Figure 6 The left side in the middle is the original white light image, and the right side is the first reconstructed image in the 590nm-620nm band. It can be seen from the figure that the first reconstructed image in the target band removes the influence of blood stains and can better show the location of the bleeding point.

[0178] In this embodiment, the above-mentioned image reconstruction method is applied to the detection of bleeding points, so that a first reconstructed image with better quality can be reconstructed more quickly, thereby improving the efficiency and accuracy of bleeding point detection.

[0179] In one embodiment, in order to facilitate those skilled in the art to understand the embodiments of the present application, as shown in FIG. Figure 7 As shown, the following also provides a specific flow chart of an image reconstruction method. In this embodiment, the method includes the following steps:

[0180] Step S710, acquiring an original image of the detection object in a preset band collected by the endoscope imaging system, and a first spectral parameter of the endoscope imaging system in the preset band;

[0181] Step S720, constructing a pseudo-inverse matrix corresponding to the first spectral parameter in a preset band;

[0182] Step S730, determining first reflectivity information of each position on the detection object in a preset band according to the color information in the original image and the pseudo-inverse matrix;

[0183] Step S740, based on the first reflectivity information, determining second reflectivity information of each position on the detection object in the target band, and determining a second spectral parameter of the endoscope imaging system in the target band;

[0184] Step S750, generating a first reconstructed image of the detection object in the target band based on the second reflectivity information and the second spectral parameter of each position;

[0185] Step S760, performing color space conversion processing on the first reconstructed image and the original image respectively to obtain a converted reconstructed image and a converted original image; the color space conversion processing is used to convert from the RGB color space to the YCC color space;

[0186] Step S770, generating a fused image based on the color value of the converted reconstructed image in the Y channel and the color values ​​of the converted original image in the three YCC channels;

[0187] Step S780: convert the fused image from the YCC color space back to the RGB color space to obtain an enhanced image corresponding to the original image.

[0188] The image reconstruction method provided in this embodiment has the following beneficial effects:

[0189] 1. No special optical imaging equipment is required, which can save hardware costs;

[0190] 2. A single image of the object to be detected is required as input, and no prior image is required;

[0191] 3. The wavelength of the image created can be accurate to 1nm;

[0192] 4. Capable of reconstructing images at any wavelength or band within the preset band range;

[0193] 5. The span of the target band can be modified according to needs, with high flexibility;

[0194] 6. Compared with the FICE system which is only applicable to white light images, the method proposed in this application can be applied to special light and has a wider application value in the wavelength band;

[0195] 7. When applied to bleeding point detection, it can remove the interference of blood stains and improve the detection efficiency of bleeding points.

[0196] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0197] Based on the same inventive concept, the embodiment of the present application also provides an image reconstruction system for implementing the above-mentioned image reconstruction method. The implementation solution provided by the system to solve the problem is similar to the implementation solution recorded in the above-mentioned method, so the specific limitations in one or more image reconstruction system embodiments provided below can refer to the limitations on the image reconstruction method above, and will not be repeated here.

[0198] In one embodiment, Figure 8 As shown, an image reconstruction system is provided, including: an acquisition unit 810, a determination unit 820 and a reconstruction unit 830, wherein:

[0199] The acquisition unit 810 is used to acquire the original image of the detection object in a preset band; the original image is acquired by an endoscope imaging system;

[0200] The determination unit 820 is used to determine the first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band;

[0201] The reconstruction unit 830 is used to generate a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter; the target band includes at least one target wavelength, and the target band is within the range of a preset band.

[0202] In one embodiment, the determination unit 820 is further used to construct a pseudo-inverse matrix corresponding to the first spectral parameter in a preset band; and determine the first reflectivity information of each position on the detection object in the preset band based on the color information in the original image and the pseudo-inverse matrix.

[0203] In one embodiment, the reconstruction unit 830 is further used to determine the second reflectivity information of each position on the detection object in the target band based on the first reflectivity information; determine the second spectral parameters of the endoscopic imaging system in the target band; and generate a first reconstructed image of the detection object in the target band based on the second reflectivity information and the second spectral parameters of each position.

[0204] In one of the embodiments, the target band includes at least two target wavelengths; the reconstruction unit 830 is further used to generate a second reconstructed image of the detection object at each target wavelength based on the first reflectivity information and the first spectral parameter; and perform image synthesis on the second reconstructed images of all target wavelengths to obtain a first reconstructed image.

[0205] In one of the embodiments, the reconstruction unit 830 is also used to determine, for any one of the target wavelengths to be processed, based on the first reflectivity information, the third reflectivity information of each position on the detection object at the wavelength to be processed; determine the third spectral parameter of the endoscopic imaging system at the wavelength to be processed; and generate a second reconstructed image of the detection object at the wavelength to be processed based on the third reflectivity information and the third spectral parameter of each position.

[0206] In one of the embodiments, the system further includes an enhancement unit for performing image enhancement processing on the original image based on the first reconstructed image to obtain an enhanced image corresponding to the original image.

[0207] In one of the embodiments, the original image and the first reconstructed image are both images in the RGB color space; the enhancement unit is further used to perform color space conversion processing on the first reconstructed image and the original image respectively to obtain a converted reconstructed image and a converted original image; the color space conversion processing is used to convert from the RGB color space to the YCC color space; based on the color value of the converted reconstructed image in the Y channel and the color values ​​of the converted original image in the three YCC channels, a fused image is generated; the fused image is converted from the YCC color space back to the RGB color space to obtain an enhanced image.

[0208] In one of the embodiments, the target wavelength band is 590nm-620nm; the system also includes a detection unit for identifying bleeding points existing in the detection object according to the first reconstructed image.

[0209] In one embodiment, the endoscopic imaging system includes: a light source, a sensor, and an optical component; the first spectral parameter includes the spectral emissivity of the light source, the spectral sensitivity distribution information of the sensor, and the spectral transmittance of the optical component.

[0210] In one embodiment, the endoscopic imaging system further includes a filter; and the first spectral parameter further includes a spectral transmittance of the filter.

[0211] Each part of the above image reconstruction system can be implemented in whole or in part by software, hardware or a combination thereof. Each part can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each part.

[0212] In one embodiment, an image processing device is provided. The internal structure diagram of the image processing device can be as follows: Fig. 9As shown. The image processing device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the image processing device is used to provide computing and control capabilities. The memory of the image processing device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the image processing device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an image reconstruction method is implemented. The display screen of the image processing device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the image processing device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the housing of the image processing device, or an external keyboard, touchpad or mouse, etc.

[0213] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the image processing device to which the scheme of the present application is applied. The specific image processing device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0214] In one embodiment, an image processing device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0215] In one embodiment, an endoscope system is provided, comprising: an endoscope, a display, a light source device, and Fig. 9 The image processing device shown can implement the steps in the above method embodiments.

[0216] The endoscope and the light source device may constitute an endoscope imaging system, and the endoscope system in this embodiment includes an endoscope imaging system. The endoscope includes a sensor and an optical component (such as a lens). The light source device includes a light source, and the light source device may also include a filter. Whether a filter is provided in the light source device may be determined according to actual conditions.

[0217] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0218] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0219] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0220] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An image reconstruction method, characterized in that: The method comprises: Acquire an original image of the detection object in a preset band; the original image is acquired by an endoscope imaging system; Determine first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band; A first reconstructed image of the detection object in a target band is generated according to the first reflectivity information and the first spectral parameter; the target band includes at least one target wavelength, and the target band is within the range of the preset band.

2. The method according to claim 1, characterized in that The endoscope imaging system comprises: a light source, a sensor and an optical component; The first spectral parameters include the spectral emissivity of the light source, the spectral sensitivity distribution information of the sensor, and the spectral transmittance of the optical component.

3. The method according to claim 2, characterized in that The endoscope imaging system further comprises: a filter; The first spectral parameter also includes the spectral transmittance of the filter.

4. The method according to any one of claims 1 to 3, characterized in that The determining, based on the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band, the first reflectivity information of each position on the detection object in the preset band includes: Constructing a pseudo-inverse matrix corresponding to the first spectral parameter in the preset band; The first reflectivity information of each position on the detection object in the preset band is determined according to the color information in the original image and the pseudo-inverse matrix.

5. The method according to any one of claims 1 to 3, characterized in that The step of generating a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter includes: Based on the first reflectivity information, determining second reflectivity information of each position on the detection object in the target band; Determining a second spectral parameter of the endoscopic imaging system in the target band; Based on the second reflectivity information of each position and the second spectral parameter, a first reconstructed image of the detection object in the target band is generated.

6. The method according to any one of claims 1 to 3, characterized in that The target wavelength band includes at least two target wavelengths; and generating a first reconstructed image of the detection object in the target wavelength band according to the first reflectivity information and the first spectral parameter includes: generating a second reconstructed image of the detection object at each of the target wavelengths according to the first reflectivity information and the first spectral parameter; The second reconstructed images of all the target wavelengths are synthesized to obtain the first reconstructed image.

7. The method according to claim 6, characterized in that Generating a second reconstructed image of the detection object at each of the target wavelengths according to the first reflectivity information and the first spectral parameter includes: For any one to-be-processed wavelength among the target wavelengths, determining third reflectivity information of each position on the detection object at the to-be-processed wavelength based on the first reflectivity information; determining a third spectral parameter of the endoscopic imaging system at the wavelength to be processed; Based on the third reflectivity information of each position and the third spectral parameter, a second reconstructed image of the detection object at the wavelength to be processed is generated.

8. The method according to any one of claims 1 to 3, characterized in that After generating a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter, the method further includes: Based on the first reconstructed image, image enhancement processing is performed on the original image to obtain an enhanced image corresponding to the original image.

9. The method according to claim 8, characterized in that The original image and the first reconstructed image are both images in RGB color space; The performing image enhancement processing on the original image based on the first reconstructed image to obtain an enhanced image corresponding to the original image includes: Performing color space conversion processing on the first reconstructed image and the original image respectively to obtain a converted reconstructed image and a converted original image; The color space conversion process is used to convert from RGB color space to YCC color space; Generate a fused image based on the color value of the converted reconstructed image in the Y channel and the color values ​​of the converted original image in the three YCC channels; The fused image is converted from the YCC color space back to the RGB color space to obtain the enhanced image.

10. The method according to any one of claims 1 to 3, characterized in that: The target wavelength is 590nm-620nm; After generating a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter, the method further includes: According to the first reconstructed image, bleeding points in the detection object are identified.

11. An image reconstruction system, characterized in that: The device comprises: An acquisition unit, used for acquiring an original image of the detection object in a preset band; the original image is acquired by an endoscope imaging system; A determination unit, configured to determine first reflectivity information of each position on the detection object in the preset band according to the color information in the original image and the first spectral parameter of the endoscope imaging system in the preset band; A reconstruction unit is used to generate a first reconstructed image of the detection object in a target band according to the first reflectivity information and the first spectral parameter; the target band includes at least one target wavelength, and the target band is within the range of the preset band.

12. An image processing device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the image enhancement method according to any one of claims 1 to 10 are implemented.

13. An endoscope system, characterized in that: include: An endoscope, a display, a light source device, and an image processing device as claimed in claim 12.

Images