Oral cavity internal image enhancement method and device under low illumination, equipment and medium

Abstract

The invention relates to the technical field of image processing, and discloses an oral cavity internal image enhancement method under low illumination, which comprises the following steps: acquiring oral cavity internal image sets with different brightness, and dividing oral cavity images into image block sets; calculating an image exposure loss value according to the image block set, and calculating a color loss value and a brightness loss value according to each color channel of the oral cavity internal image; performing parameter optimization on a preset high-order curve model according to the image exposure loss value, the color loss value and the brightness loss value to obtain an optimized curve model; calculating a pixel mapping relation of the to-be-enhanced oral cavity internal image according to the optimized curve model; and performing pixel mapping processing on the to-be-enhanced oral cavity internal image by using the pixel mapping relation to obtain an enhanced oral cavity image. The invention further provides an oral cavity internal image enhancement device under low illumination, computer equipment and a storage medium. The image quality of the oral cavity internal image can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular, to an oral internal image enhancement method, device, equipment and medium under low light. Background Art

[0002] When using an oral scanning device to scan and model teeth and gums, the original images obtained by the device usually have the problem of insufficient brightness. Therefore, an oral internal image enhancement method is needed to restore the image brightness and restore the true condition of the oral cavity. Images with low light cannot accurately reflect image information, which not only affects the user's scanning perception experience but also transmits non-ideal information to the subsequent processing of the system. At the same time, due to the complex internal environment of the oral cavity, if the brightness of the fill light on the scanner is increased, overexposure will occur for metal materials such as implant posts or saliva bubbles. Therefore, during the actual use of the oral scanning device, the active fill light brightness of the device will be relatively low, and then an image enhancement method is used to restore the image brightness.

[0003] Traditional image enhancement methods usually simply enhance the brightness in the HSV color space. On the other hand, there is also the method of histogram equalization, which aims to enhance the brightness by rebalancing the gray distribution of the image. Although these traditional methods are simple and have high computational efficiency, the algorithms have poor robustness, are sensitive to noise, and have great limitations in the usage scenarios, and cannot achieve ideal results in some cases. Another common type of image enhancement algorithm is the Retinex-based image enhancement algorithm, which mimics the contrast perception of the human visual system, separates the color and illumination components of the image based on the assumptions of color constancy and spatial separation, and improves the image brightness by enhancing the reflection. However, the Retinex algorithm still has limitations in the usage scenarios. With the development of deep learning, some deep learning-based methods for solving image enhancement problems have gradually emerged. The more common methods include the network structure based on CNN, which mainly realizes the enhancement of image brightness by learning the mapping relationship between low-light images and normal images; the network structure based on GAN, which improves the image quality through the generation confrontation from low light to normal images. These deep learning-based methods have stronger adaptability and better enhancement effects in various complex scenarios, but they have a large amount of calculation, complex network structures, are difficult to meet the requirements of real-time enhancement, and most of them require corresponding low-light images and normal images as training data.

[0004] Therefore, how to improve the image quality of oral internal images has become an urgent problem to be solved. Summary of the Invention

[0005] The present invention provides a method, device, equipment and medium for enhancing the internal oral image under low light, and its main purpose is to solve the problem of poor image quality of the existing internal oral images.

[0006] To achieve the above object, a method for enhancing the internal oral image under low light provided by the present invention includes:

[0007] Obtain a set of internal oral images with different brightnesses, and divide each internal oral image in the set of internal oral images into a set of image blocks;

[0008] Calculate the image exposure loss value according to the set of image blocks, and calculate the color loss value and the brightness loss value according to the RGB color channels of the internal oral image;

[0009] Optimize the parameters of a preset high-order curve model according to the image exposure loss value, the color loss value and the brightness loss value to obtain an optimized curve model;

[0010] Calculate the pixel mapping relationship of the internal oral image to be enhanced according to the optimized curve model;

[0011] Perform pixel mapping processing on the internal oral image to be enhanced by using the pixel mapping relationship to obtain an enhanced oral image.

[0012] Optionally, the calculating the image exposure loss value according to the set of image blocks includes:

[0013] Calculate the average gray value of each image block in the set of image blocks;

[0014] Calculate the image exposure loss value corresponding to the internal oral image according to the average gray value;

[0015] Calculate the image exposure loss value by using the following formula:

[0016]

[0017] where, L ex represents the image exposure loss value, N represents the total number of image blocks, M k represents the average gray value of the kth image block, and E represents a preset exposure constant.

[0018] Optionally, the calculating the color loss value and the brightness loss value according to the RGB color channels of the internal oral image includes:

[0019] Calculate the average pixel value and the directional gradient of the internal oral image in the RGB color channels respectively;

[0020] Calculate the color loss value corresponding to the internal oral image according to the average pixel value;

[0021] Calculate the color loss value corresponding to the internal oral image using the following formula:

[0022] L col = ∑ (p,q)∈ε (J p - J q ) 2 , ε = {(R, G), (R, B), (B, G)}

[0023] Wherein, L col represents the color loss value, J p represents the average pixel value of the p-th RGB color channel, J q represents the average pixel value of the q-th RGB color channel, and R, G, and B respectively represent the RGB color channels;

[0024] Calculate the brightness loss value corresponding to the oral image according to the direction gradient.

[0025] Optionally, the calculating the brightness loss value corresponding to the internal oral image according to the direction gradient includes:

[0026] Calculate the brightness loss value corresponding to the oral image using the following formula:

[0027]

[0028] Wherein, L(ill) represents the brightness loss value, represents the direction gradient in the horizontal direction, represents the direction gradient in the vertical direction, represents the parameter matrix of the c color channel during the f-th parameter optimization, F represents the total number of parameter optimizations, and R, G, and B represent the RGB color channels.

[0029] Optionally, the optimizing the model parameters of the preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model includes:

[0030] Take the sum of the image exposure loss value, the color loss value, and the brightness loss value as the target loss value;

[0031] Use the preset CNN network model to optimize the model parameters of the high-order curve model according to the target loss value until the target loss value is less than the preset loss value threshold, and obtain the curve model parameters corresponding to the high-order curve model;

[0032] Perform parameter replacement on the model parameters according to the curve model parameters to obtain an optimized curve model.

[0033] Optionally, optimizing the parameters of a preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain curve model parameters includes:

[0034] The high-order curve model is shown as follows:

[0035] IE n (x) = IE n-1 (x) + α n IE n-1 (x)(1 - IE n-1 (x))

[0036] Wherein, IE n (x) represents the intraoral image corresponding to the nth parameter optimization, and IE n-1 (x) represents the intraoral image corresponding to the (n - 1)th iteration, and α n represents the model parameter of the high-order curve model after the nth iteration.

[0037] Optionally, calculating the pixel mapping relationship of the intraoral image to be enhanced according to the optimized curve model includes:

[0038] Performing pixel mapping processing on the intraoral image to be enhanced for a preset number of iterations by using the optimized curve model to obtain the iteration mapping relationship corresponding to each iteration;

[0039] Selecting the last iteration mapping relationship in the iteration mapping relationships as the pixel mapping relationship of the intraoral image to be enhanced.

[0040] To solve the above problems, the present invention further provides an intraoral image enhancement device under low light. The device includes:

[0041] An image block set partitioning module, configured to obtain an intraoral image set with different brightnesses, and partition each intraoral image in the intraoral image set into an image block set;

[0042] An image loss value calculation module, configured to calculate an image exposure loss value according to the image block set, and calculate a color loss value and a brightness loss value according to the RGB color channels of the intraoral image;

[0043] A parameter optimization module, configured to optimize the parameters of a preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model;

[0044] A pixel mapping relationship calculation module, configured to calculate the pixel mapping relationship of the intraoral image to be enhanced according to the optimized curve model;

[0045] An image enhancement module, configured to perform pixel mapping processing on the oral internal image to be enhanced by using the pixel mapping relationship, so as to obtain an enhanced oral image.

[0046] To solve the above problems, the present invention further provides a computer device, which includes:

[0047] At least one processor;

[0048] And a memory communicatively connected to the at least one processor;

[0049] Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute the above-mentioned method for enhancing an oral internal image under low light.

[0050] To solve the above problems, the present invention further provides a computer-readable storage medium, in which at least one computer program is stored, and the at least one computer program is executed by a processor in a computer device to implement the above-mentioned method for enhancing an oral internal image under low light.

[0051] In an embodiment of the present invention, an oral internal image set with different brightnesses is collected, each oral internal image is divided into image blocks to obtain an image block set; an exposure loss value is calculated according to the image block set, and a color loss value and a brightness loss value are calculated according to each color channel; the parameters of a preset high-order curve model are optimized according to the exposure loss value, the color loss value and the brightness loss value to obtain an optimized curve model; a pixel mapping relationship of the oral internal image to be enhanced is calculated according to the optimized curve model; and pixel mapping processing is performed on the oral internal image to be enhanced according to the pixel mapping relationship to obtain an enhanced oral image. The image brightness of the oral internal image to be enhanced can be enhanced, the detail information is completely retained, a more accurate enhanced oral image is obtained, and the image quality of the oral internal image is improved. Therefore, the method, device, computer device and computer-readable storage medium for enhancing an oral internal image under low light proposed by the present invention can solve the problem of poor image quality of existing oral internal images. Description of the Drawings

[0052] Figure 1 It is a schematic flowchart of a method for enhancing an oral internal image under low light provided by an embodiment of the present invention;

[0053] Figure 2 It is a schematic diagram of the effect after image enhancement of an oral internal image to be enhanced provided by an embodiment of the present invention;

[0054] Figure 3Functional module diagram of the oral cavity internal image enhancement device under low light provided by an embodiment of the present invention;

[0055] Figure 4 Schematic structural diagram of an electronic device for implementing the oral cavity internal image enhancement method under low light provided by an embodiment of the present invention.

[0056] The realization, functional characteristics and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0057] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0058] An embodiment of the present application provides an oral cavity internal image enhancement method under low light. The execution subject of the oral cavity internal image enhancement method under low light includes, but is not limited to, at least one of computer devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the oral cavity internal image enhancement method under low light can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, Content Delivery Network (CDN), and big data and artificial intelligence platforms.

[0059] Refer to Figure 1 As shown, it is a flowchart of the oral cavity internal image enhancement method under low light provided by an embodiment of the present invention. In this embodiment, the oral cavity internal image enhancement method under low light includes:

[0060] S1. Obtain an oral cavity internal image set with different brightnesses, and divide each oral cavity internal image in the oral cavity internal image set into an image block set.

[0061] In an embodiment of the present invention, the oral cavity internal image set is an oral cavity internal image scanned by an oral cavity scanning device under different brightnesses;

[0062] Divide each oral cavity image into equal image blocks. For example, it can be divided into image blocks of 32×32 size, or a preset number of image blocks, to obtain an image block set.

[0063] S2. Calculate the image exposure loss value according to the set of image blocks, and calculate the color loss value and the brightness loss value according to the RGB color channels of the intraoral image.

[0064] Furthermore, the image exposure loss value can adjust the exposure intensity to prevent overexposure or underexposure. The exposure intensity can be controlled through the image exposure loss value, and the gray value of each image block can be controlled within a certain range.

[0065] In the embodiment of the present invention, calculating the image exposure loss value according to the set of image blocks includes:

[0066] Calculate the average gray value of each image block in the set of image blocks;

[0067] Calculate the image exposure loss value corresponding to the intraoral image according to the average gray value.

[0068] Specifically, the image exposure loss value is calculated using the following formula:

[0069]

[0070] where L ex represents the image exposure loss value, N represents the total number of image blocks, M k represents the average gray value of the k-th image block, and E represents a preset exposure constant.

[0071] In the embodiment of the present invention, the color loss value represents the degree of deviation of the color space of the intraoral image, and the brightness loss value represents the degree of reduction or distortion of the brightness information of the intraoral image. The color loss value and the brightness loss value can characterize the image quality of the intraoral image from both the color and brightness aspects.

[0072] In the embodiment of the present invention, calculating the color loss value and the brightness loss value according to the RGB color channels of the intraoral image includes:

[0073] Calculate the average pixel value and the directional gradient of the intraoral image in the RGB color channels respectively;

[0074] Calculate the color loss value corresponding to the intraoral image according to the average pixel value;

[0075] Calculate the brightness loss value corresponding to the intraoral image according to the directional gradient.

[0076] In the embodiment of the present invention, the Euclidean distance between two RGB color channels can be calculated through the average pixel value of the color channels, thereby reflecting the degree of deviation of the color space of the intraoral image.

[0077] Specifically, the color loss value corresponding to the oral internal image is calculated using the following formula:

[0078] L col = ∑ (p,q)∈ε (J p - J q ) 2 , ε = {(R, G), (R, B), (B, G)}

[0079] where L col represents the color loss value, J p represents the average pixel value of the p-th RGB color channel, and J q represents the average pixel value of the q-th RGB color channel. R, G, and B represent the RGB color channels.

[0080] Furthermore, the directional gradient is the gradient of the R, G, and B color channels in the x (horizontal) and y (vertical) directions. The gradients of the color channels in the x (horizontal) and y (vertical) directions can be obtained by calculating the differences between adjacent pixels. For example, by calculating the difference between the current pixel and its adjacent pixel on the right in the x direction, the gradient in the x direction is obtained, and by calculating the difference between the current pixel and its adjacent pixel below in the y direction, the gradient in the y direction is obtained.

[0081] Specifically, the brightness loss value corresponding to the oral internal image is calculated using the following formula:

[0082]

[0083] where L(ill) represents the brightness loss value, represents the directional gradient in the horizontal direction, represents the directional gradient in the vertical direction, represents the parameter matrix of the c color channel during the f-th parameter optimization, F represents the total number of parameter optimizations, and R, G, and B represent the RGB color channels.

[0084] In the embodiments of the present invention, by calculating the color loss value and the brightness loss value, the degree of reduction or distortion of the brightness information and the color space deviation of the oral internal image can be represented, so as to more comprehensively evaluate the image quality of the oral internal image and improve the effect of subsequent image enhancement.

[0085] S3. Optimize the parameters of the preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model.

[0086] In the embodiments of the present invention, the sum of three losses, namely the image exposure loss value, the color loss value, and the brightness loss value, is used as the loss function of the CNN network model for training to obtain more accurate curve model parameters.

[0087] In the embodiments of the present invention, the curve model parameters are the optimized parameters corresponding to the high-order curve model. Through the optimized parameters, the optimal enhanced image can be obtained, thereby improving the effect of oral internal image enhancement.

[0088] In the embodiments of the present invention, the parameter optimization of the preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model includes:

[0089] Taking the sum of the image exposure loss value, the color loss value, and the brightness loss value as the target loss value;

[0090] According to the target loss value, using the preset CNN network model to optimize the model parameters of the high-order curve model until the target loss value is less than the preset loss value threshold to obtain the curve model parameters corresponding to the high-order curve model;

[0091] Performing parameter replacement on the model parameters according to the curve model parameters to obtain an optimized curve model.

[0092] In one embodiment, the CNN (Convolutional Neural Network) network model is a convolutional neural network. The convolutional neural network is composed of 6 convolutional blocks, and each convolutional block consists of a 3×3 convolutional kernel and an activation function. Through the convolutional layers composed of multiple 3×3 convolutional kernels and an activation function in the CNN network model, multiple convolutional features are obtained; performing skip connections on the multiple convolutional features is beneficial for information interaction and fusion of convolutional features in different layers, and is conducive to adaptively learning the features of each color channel; using a convolutional layer and an activation layer as the output layer for the multiple convolutional features, the curve model parameters can be learned through the CNN network model to obtain the curve model parameters.

[0093] Specifically, the oral internal image corresponding to each parameter optimization has a mapping relationship with the output of the previous time, and the model parameters of the high-order curve model can be obtained each time optimization is performed. For example:

[0094] IE(x) = I(x) + αI(x)(1 - I(x))

[0095] IE 2 (x) = IE(x) + α 2 IE(x)(1 - IE(x))

[0096] Among them, IE(x) represents the intraoral image corresponding to the first parameter optimization, I(x) represents the intraoral image, α represents the model parameters of the first iteration, and IE 2 (x) represents the intraoral image corresponding to the second parameter optimization, and α 2 represents the model parameters of the second iteration.

[0097] Furthermore, the model parameters of the intraoral image at the nth iteration can be obtained as shown in the following formula:

[0098] IE n (x) = IE n-1 (x) + α n IE n-1 (x)(1 - IE n-1 (x))

[0099] Among them, IE n (x) represents the intraoral image corresponding to the nth parameter optimization, and IE n-1 (x) represents the intraoral image corresponding to the (n - 1)th iteration, and α n represents the model parameters of the high-order curve model after the nth iteration.

[0100] In the embodiments of the present invention, each pixel point in the intraoral image corresponds to a high-order curve model, that is, each different α n has a different mapping relationship, and high-order curve models with different shapes can be obtained. Furthermore, pixel mapping can be performed on each pixel in the intraoral image to be enhanced to obtain an enhanced oral image, effectively improving the image enhancement effect of the intraoral image to be enhanced.

[0101] S4. Calculate the pixel mapping relationship of the intraoral image to be enhanced according to the optimized curve model

[0102] In the embodiments of the present invention, the pixel values of the pixel points in the intraoral image to be enhanced are mapped through the model parameters of the nth iteration to obtain the target oral image.

[0103] Specifically, calculating the pixel mapping relationship of the intraoral image to be enhanced according to the optimized curve model includes:

[0104] Performing pixel mapping processing on the intraoral image to be enhanced for a preset number of iterations by using the optimized curve model to obtain the iteration mapping relationship corresponding to each iteration;

[0105] Selecting the last iteration mapping relationship in the iteration mapping relationships as the pixel mapping relationship of the intraoral image to be enhanced.

[0106] In the embodiment of the present invention, the model parameters of the high-order curve model in the last iteration mapping relationship are used as the pixel mapping relationship, and each pixel point can be iteratively mapped multiple times.

[0107] S5. Perform pixel mapping processing on the to-be-enhanced intraoral image by using the pixel mapping relationship to obtain an enhanced oral image.

[0108] Specifically, the following formula is used to perform pixel mapping processing on the to-be-enhanced intraoral image:

[0109]

[0110] Wherein, represents the enhanced oral image, I(x) represents the to-be-enhanced intraoral image, and α n represents the pixel mapping relationship.

[0111] In the embodiment of the present invention, for the effect after image enhancement of the to-be-enhanced intraoral image, refer to Figure 2 as shown. Among them, Figure 2 (a) in represents the to-be-enhanced intraoral image, and (b) represents the enhanced oral image corresponding to the to-be-enhanced intraoral image. By performing image enhancement on the to-be-enhanced intraoral image, the image brightness of the to-be-enhanced intraoral image can be enhanced, the image information can be clearly restored, the detail information can be completely retained, a more accurate target oral image can be obtained, and the image quality of the intraoral image can be improved.

[0112] As Figure 3 shown, it is a functional module diagram of an intraoral image enhancement device under low light provided by an embodiment of the present invention.

[0113] The intraoral image enhancement device 100 under low light according to the present invention can be installed in a computer device. According to the implemented functions, the intraoral image enhancement device 100 under low light can include an image block set division module 101, an image loss value calculation module 102, parameter optimization 103, a pixel mapping relationship calculation module 104, and an image enhancement module 105. The modules in the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of a computer device and can complete fixed functions, and are stored in the memory of the computer device.

[0114] In this embodiment, the functions of each module / unit are as follows:

[0115] The image block set division module 101 is used to obtain an intraoral image set with different brightnesses, and divide each intraoral image in the intraoral image set into an image block set;

[0116] The image loss value calculation module 102 is configured to calculate an image exposure loss value according to the set of image blocks, and calculate a color loss value and a brightness loss value according to the RGB color channels of the intraoral image;

[0117] The parameter optimization module 103 is configured to optimize the parameters of a preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model;

[0118] The pixel mapping relationship calculation module 104 is configured to calculate the pixel mapping relationship of the intraoral image to be enhanced according to the optimized curve model;

[0119] The image enhancement module 105 is configured to perform pixel mapping processing on the intraoral image to be enhanced by using the pixel mapping relationship to obtain an enhanced oral image.

[0120] Specifically, each module in the intraoral image enhancement device 100 under low light in the embodiments of the present invention adopts the same technical means as those in the above Figures 1 to 2 intraoral image enhancement method under low light, and can produce the same technical effects, which will not be elaborated here.

[0121] As Figure 4 shown, it is a schematic structural diagram of an electronic device for implementing the intraoral image enhancement method under low light of teeth provided by an embodiment of the present invention.

[0122] The electronic device 10 may include a processor 11, a memory 12, a communication bus 13, and a communication interface 14, and may further include a computer program stored in the memory 12 and executable on the processor 11, such as a dental arch three-dimensional digital model segmentation method program.

[0123] Among them, the processor 11 may be composed of integrated circuits in some embodiments. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same or different functions, including a combination of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips.

[0124] The memory 12 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as: SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. The memory 102 may be an internal storage unit of the electronic device in some embodiments, such as the mobile hard disk of the electronic device.

[0125] The communication bus 13 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is set to implement the connection and communication between the memory 12 and at least one processor 11, etc.

[0126] The communication interface 14 is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), and is generally used to establish a communication connection between this electronic device and other electronic devices.

[0127] Only the electronic device with components is shown in the figure. Those skilled in the art can understand that the structure shown in the figure does not constitute a limitation on the electronic device, and it may include fewer or more components than shown in the figure, or combine certain components, or have different component arrangements.

[0128] It should be understood that the above embodiments are only for illustration purposes and are not limited by this structure in the scope of the patent application.

[0129] Furthermore, if the modules / units integrated in the electronic device 10 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM, Read-Only Memory).

[0130] The present invention also provides a computer-readable storage medium. The readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, it can implement:

[0131] Obtain a set of oral internal images with different brightnesses, and divide each oral internal image in the set of oral internal images into a set of image blocks;

[0132] Calculate an image exposure loss value according to the set of image blocks, and calculate a color loss value and a brightness loss value according to the RGB color channels of the oral internal image;

[0133] Optimize the parameters of a preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model;

[0134] Calculate the pixel mapping relationship of the oral internal image to be enhanced according to the optimized curve model;

[0135] Perform pixel mapping processing on the oral internal image to be enhanced by using the pixel mapping relationship to obtain an enhanced oral image.

[0136] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0137] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0138] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any associated drawing marks in the claims should not be regarded as limiting the claimed rights.

[0139] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is a theory, method, technology, and application device that uses a digital computer or a machine controlled by a digital computer to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0140] In addition, obviously, the word "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or apparatuses stated in the apparatus claims can also be implemented by one unit or apparatus through software or hardware. The terms first, second, etc. are used to represent names and do not represent any specific order.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for enhancing an intraoral image under low light, characterized in that: The method comprises: Acquire a set of intra-oral images of different brightness, and divide each intra-oral image in the set of intra-oral images into a set of image blocks; Calculating an image exposure loss value according to the image block set, and calculating a color loss value and a brightness loss value according to the RGB color channels of the intra-oral image; Optimizing the parameters of a preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model; Calculating the pixel mapping relationship of the intra-oral image to be enhanced according to the optimization curve model; The pixel mapping relationship is used to perform pixel mapping processing on the intra-oral image to be enhanced to obtain an enhanced oral image.

2. The method for enhancing the intraoral image under low light conditions according to claim 1, wherein: The calculating the image exposure loss value according to the image block set includes: Calculating the mean grayscale value of each image block in the image block set; Calculating the image exposure loss value corresponding to the intra-oral image according to the gray value mean; The image exposure loss value is calculated using the following formula: Among them, L ex represents the image exposure loss value, N represents the total number of image blocks, M k represents the mean gray value of the kth image block, and E represents the preset exposure constant.

3. The method for enhancing the intraoral image under low light conditions according to claim 1, wherein: Calculating a color loss value and a brightness loss value according to the RGB color channels of the intra-oral image includes: Calculate the average pixel value and directional gradient of the intra-oral image in the RGB color channels respectively; Calculating a color loss value corresponding to the intra-oral image according to the average pixel value; The color loss value corresponding to the intraoral image is calculated using the following formula: L col =∑ (p,q)∈ε (J p -J q ) 2 ,ε={(R,G),(R,B),(B,G)} Among them, L col Represents the color loss value, J p represents the average pixel value of the pth RGB color channel, J q represents the average pixel value of the qth RGB color channel, R, G, and B represent the RGB color channels respectively; The brightness loss value corresponding to the intra-oral image is calculated according to the directional gradient.

4. The method for enhancing the intraoral image under low light conditions according to claim 3, wherein: The calculating the brightness loss value corresponding to the intra-oral image according to the directional gradient includes: The brightness loss value corresponding to the intraoral image is calculated using the following formula: Where L(ill) represents the brightness loss value, represents the directional gradient in the horizontal direction, represents the directional gradient in the vertical direction, represents the parameter matrix of the c color channel when the enhanced image model performs parameter optimization for the fth time, F represents the total number of parameter optimizations, and R, G, and B represent RGB color channels.

5. The method for enhancing the intraoral image under low light conditions according to claim 1, wherein: The step of optimizing parameters of a preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain an optimized curve model includes: Taking the sum of the image exposure loss value, the color loss value and the brightness loss value as a target loss value; According to the target loss value, the model parameters of the high-order curve model are optimized using a preset CNN network model until the target loss value is less than a preset loss value threshold, thereby obtaining the curve model parameters corresponding to the high-order curve model; The model parameters are replaced according to the curve model parameters to obtain an optimized curve model.

6. The method for enhancing the intraoral image under low light conditions according to claim 1, wherein: The step of optimizing parameters of a preset high-order curve model according to the image exposure loss value, the color loss value, and the brightness loss value to obtain curve model parameters includes: The high-order curve model is shown below: IE n (x)=IE n-1 (x)+α n IE n-1 (x)(1-IE) n-1 (x)) Among them, IE n (x) represents the corresponding oral cavity image after the nth parameter optimization, IE n-1 (x) represents the corresponding oral cavity image after the n-1th iteration, α n Represents the model parameters of the higher-order curve model after the nth iteration.

7. The method for enhancing the intraoral image under low light conditions according to claim 1, wherein: The step of calculating the pixel mapping relationship of the oral cavity internal image to be enhanced according to the optimization curve model includes: Using the optimization curve model, pixel mapping is performed on the intra-oral image to be enhanced for a preset number of iterations to obtain an iterative mapping relationship corresponding to each iteration; The last iterative mapping relationship in the iterative mapping relationships is selected as the pixel mapping relationship of the intra-oral image to be enhanced.

8. An intraoral image enhancement device under low light, characterized in that: The device comprises: An image block set division module is used to obtain an intra-oral image set of different brightness, and divide each intra-oral image in the intra-oral image set into an image block set; An image loss value calculation module, used to calculate an image exposure loss value according to the image block set, and calculate a color loss value and a brightness loss value according to the RGB color channels of the intra-oral image; A parameter optimization module, used to optimize the parameters of a preset high-order curve model according to the image exposure loss value, the color loss value and the brightness loss value to obtain an optimized curve model; A pixel mapping relationship calculation module, used for calculating the pixel mapping relationship of the intra-oral image to be enhanced according to the optimization curve model; The image enhancement module is used to perform pixel mapping processing on the internal oral cavity image to be enhanced by using the pixel mapping relationship to obtain an enhanced oral cavity image.

9. A computer device, characterized in that: The computer device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for intra-oral image enhancement under low light as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for enhancing intra-oral images under low light conditions as described in any one of claims 1 to 7 is implemented.

Images