Adaptive enhancement method for digital ray 16-bit latent image data details
Through logarithmic function mapping and grayscale model construction, the fusion coefficient is calculated and iteratively enhanced, the problem of poor detail enhancement effect of digital ray 16-bit latent image data is solved, and higher stability and universality are achieved.
Patent Information
- Application Number
- CN202411482845.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-23
- Publication Date
- 2025-02-28
AI Technical Summary
The prior art is difficult to effectively solve the problem of enhanced details of digital ray 16-bit latent image data, resulting in problems such as submerged small defects, weak defect characteristics, unclear defect edges and hazy imaging, and low stability and poor universality.
The logarithmic function is used to map the 16-bit latent image data of the digital ray to the logarithmic space, and the floating-point number limit matrix and the Gaussian fuzzy limit matrix are constructed through the grayscale scattering model and the grayscale noise model, and the indirect main fusion coefficient and fine-tuned fusion coefficient are calculated, detail enhancement is achieved through multiple iterations, and the logarithmic space fusion result is restored to the original space through normalization method.
It effectively improves the detailed enhancement effect of digital ray 16-bit latent image data, improves stability and universality, and solves the problems of submerged small defects and unclear defect edges.
Smart Images
Figure CN119540084A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the technical field of image data processing, and in particular to a method for adaptively enhancing details of 16-bit latent image data of digital rays. Background Art
[0002] Digital radiographic flaw detection mainly uses the information of radiation intensity attenuation to characterize the properties, state and structure of the object being tested. It uses a computer to convert the digital radiation latent image data matrix into an image observable to the naked eye based on the latent image data corresponding to the intensity distribution formed after the radiation penetrates the object being tested. This not only obtains an intuitive image display of the interior of the object being tested, but also can quickly and accurately identify internal defects of the object.
[0003] In the process of digital radiographic flaw detection, the latent image data formed after the rays penetrate the object to be detected is mainly saved as 16-bit grayscale data, and relies on detail enhancement technology for defect analysis and assessment. The detail enhancement effect of latent image data directly affects the final result of defect analysis and assessment. Detail enhancement is an indispensable part of industrial flaw detection using digital rays. Due to the joint constraints of detector performance, on-site interference, and stability of the ray transmitter, the edges of defects are unclear and small defects are submerged, resulting in misjudgment and misjudgment of defect assessment results. At present, the commonly used technologies to solve such problems include histogram equalization technology, Laplace sharpening technology, Gaussian Laplace sharpening technology, gamma transform technology, and wavelet transform technology. However, these technologies are difficult to effectively solve the problem of detail enhancement of small defects, and there are limitations to varying degrees in the actual application process.
[0004] At present, there are still few studies on detail enhancement of 16-bit latent image data of digital ray, which mainly applies 8-bit image processing technology, such as histogram equalization technology, Laplace sharpening technology, Gaussian Laplace sharpening technology, gamma transform technology and wavelet transform technology, to the detail enhancement process of 16-bit latent image data. Since the 16-bit latent image data display process depends on the window position and window width, it is very different from the 8-bit image display principle, which makes it difficult for 8-bit image processing technology to adapt to 16-bit latent image data. As a result, in most cases, it fails to effectively solve the problems of small defects being submerged, defect features being weakened, defect edges being unclear, and image blurring, resulting in poor detail enhancement effect of 16-bit latent image data. At the same time, low stability and poor universality are the main drawbacks of existing latent image data enhancement technology. Summary of the invention
[0005] In order to overcome the technical defects of poor detail enhancement effect, low stability and poor universality of using the existing latent image data enhancement technology to process 16-bit latent image data, an adaptive enhancement method for details of digital ray 16-bit latent image data is provided in the embodiment of the present application, and its technical solution is as follows:
[0006] S1. Grayscale scattering model construction:
[0007] The 16-bit latent image data of digital rays is mapped to the logarithmic space using a logarithmic function, and the logarithmic function value is restricted according to the requirement that the independent variable of the logarithmic function is not 0, so as to obtain a floating point restriction matrix, namely the grayscale scattering model;
[0008] S2. Grayscale noise model construction:
[0009] The 16-bit latent image data of digital rays is transformed by using Gaussian blur function, and the value of Gaussian blur function is restricted according to the requirement that the independent variable of the logarithmic function in the Gaussian blur function is not 0, so as to obtain the Gaussian blur restriction matrix, i.e., the grayscale noise model;
[0010] S3, adaptive fusion coefficient calculation:
[0011] The indirect main fusion coefficient and the fine-tuning fusion coefficient are calculated by using the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating point restriction matrix;
[0012] S4. Calculation of logarithmic space fusion results:
[0013] The fusion result of the i-th iteration, i.e., the logarithmic space fusion result, is calculated by using the difference and sum of the logarithmic function values of the corresponding elements in the indirect main fusion coefficient, the fine-tuning fusion coefficient, the Gaussian blur restriction matrix, and the floating-point restriction matrix.
[0014] Optionally, in a possible implementation of the above adaptive enhancement method,
[0015] The indirect main fusion coefficient is calculated by combining the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating point restriction matrix, the indirect fusion constant and the direct fusion constant;
[0016] Among them, the indirect fusion constant and the direct fusion constant are both predetermined based on actual data and grid search.
[0017] Optionally, in a possible implementation of the above adaptive enhancement method,
[0018] The fine-tuning fusion coefficient is calculated by combining the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating-point restriction matrix, the fine-tuning constant and the fine-tuning polynomial order;
[0019] Among them, the fine-tuning constant and the fine-tuning polynomial order are both predetermined based on actual data and grid search.
[0020] Optionally, in a possible implementation of the above adaptive enhancement method,
[0021] In S1, the floating point limit matrix is denoted as sm (x,y), the calculation formula is:
[0022] In the formula, s ε is the limit value of the logarithmic function value, x and y represent the row and column in the matrix respectively;
[0023] In S2, the Gaussian blur function is denoted as g(x,y), and its calculation formula is:
[0024] In the formula, δ i represents the Gaussian blur parameter of the i-th iteration process; sigma(ksize)=0.15(ksize-1)+0.5, δ i =sigma(2i+1), sigma and ksize represent the sign of the Gaussian blur parameter calculation function and the Gaussian kernel size of the i-th iteration process;
[0025] Correspondingly, the Gaussian fuzzy constraint matrix is denoted as g m (x,y), the calculation formula is:
[0026] In the formula, g ε The limit value of the Gaussian blur function.
[0027] Optionally, in a possible implementation of the above adaptive enhancement method,
[0028] The indirect main fusion coefficient is denoted as f β (x,y), the calculation formula is:
[0029] In the formula, f α (x,y) is the direct main fusion coefficient, and β is the indirect fusion constant.
[0030] Furthermore, the calculation formula of the direct main fusion coefficient is:
[0031] Where α is the direct fusion constant, l Δ (x, y) is the difference between the logarithmic function values of the corresponding elements in the floating point limit matrix and the Gaussian blur limit matrix.
[0032] Optionally, in a possible implementation of the above adaptive enhancement method,
[0033] The fine-tuning fusion coefficient is denoted as f γ (x,y), the calculation formula is:
[0034] In the formula, γ is the fine-tuning constant, N is the order of the fine-tuning polynomial, is the jth order of the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating point restriction matrix;
[0035] Optionally, in a possible implementation of the above adaptive enhancement method,
[0036] The fusion result of the i-th iteration is Δ i (x,y), the calculation formula is:
[0037] Δ i (x,y)=f β (x,y)l Δ (x,y)+f β (x,y)l Φ (x,y)f γ (x,y), where l Φ (x,y)=l g (x,y)+l s (x, y) is the sum of the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating-point restriction matrix.
[0038] Another aspect of an embodiment of the present application provides a computer device, including:
[0039] at least one processor; and a memory in communication with the at least one processor;
[0040] Wherein: the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method as described in any of the above items.
[0041] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores computer instructions. When the computer instructions are executed by a processor, the adaptive enhancement method as described in any of the above items is implemented.
[0042] A computer program product is also provided in an embodiment of the present application, including a computer program, which, when executed by a processor, implements the adaptive enhancement method as described in any of the above items.
[0043] The technical scheme adopted in the embodiment of the present application can achieve the following technical effects: using a logarithmic function to represent light reflection, taking the floating-point limit matrix and the Gaussian blur limit matrix of the 16-bit latent image data of the digital ray as fusion sources, using the logarithmic function to calculate the sum and difference of the corresponding element values of the two matrices, and then using variable Gaussian parameters to approximate the multi-scale transformation process, realizing the detail enhancement process through multiple iterations, and the Gaussian parameters are different in each iteration process, and the indirect main fusion coefficient and the fine-tuning fusion coefficient corresponding to the 16-bit latent image data of the digital ray are adaptively calculated according to the grayscale noise model and the grayscale scattering model, and the logarithmic space fusion result is restored to the original space by using the normalization method, thereby solving the technical defects of the existing latent image data enhancement technology in processing 16-bit latent image data, such as poor detail enhancement effect, low stability and poor universality. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The accompanying drawings exemplarily illustrate the embodiments and constitute a part of the specification, and together with the text description of the specification, are used to explain the exemplary implementation of the embodiments. The embodiments shown are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements.
[0045] Figure 1 A schematic flow chart of a method for adaptively enhancing details of 16-bit latent image data of digital rays in this application;
[0046] Figure 2 This is a schematic diagram of data fusion for the i-th iteration of the adaptive detail enhancement process of digital ray 16-bit latent image data in this application;
[0047] Figure 3 This is a schematic diagram comparing the effects of the adaptive enhancement method and the histogram method after enhancement in this application;
[0048] Figure 4 A hardware structure diagram of the computer device in this application. DETAILED DESCRIPTION
[0049] In order to make the purpose, technical solutions 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 intended to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.
[0050] It should be noted that the descriptions involving "first", "second", etc. in the embodiments of the present application are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.
[0051] In the description of the present application, it should be understood that the numerical labels before the steps do not indicate the order in which the steps are executed, but are only used to facilitate the description of the present application and to distinguish each step, and therefore should not be understood as a limitation on the present application.
[0052] Aiming at the problem of detail enhancement of 16-bit latent image data in the process of digital radiographic flaw detection and assessment, an adaptive enhancement method for the details of 16-bit latent image data of digital radiography is designed. This technical solution takes logarithmic domain operation, multi-operation and multi-scale fusion as the main lines, uses Gaussian function and logarithmic log function to establish grayscale noise model and grayscale scattering model respectively, uses changing Gaussian parameters to approximate the variable scale process, adaptively calculates the fusion coefficient of 16-bit latent image data according to the grayscale noise model and grayscale scattering model, and uses the normalization method to restore the logarithmic space fusion result to the original space, thus solving the technical defects of the existing latent image data enhancement technology in processing 16-bit latent image data, such as poor detail enhancement effect, low stability and poor universality.
[0053] See also Figure 1 and 2 In an embodiment of the present application, a method for adaptively enhancing details of 16-bit latent image data of digital rays comprises the following steps:
[0054] S1. Grayscale scattering model construction:
[0055] The 16-bit latent image data of digital rays are mapped to the logarithmic space using a logarithmic function, and the logarithmic function value is restricted according to the requirement that the independent variable of the logarithmic function is not 0, so as to obtain a floating point restriction matrix, namely the grayscale scattering model.
[0056] Specifically, the digital ray 16-bit latent image data is the latent image data corresponding to the intensity distribution formed after the digital ray penetrates the object to be inspected during the digital ray flaw detection process and stored in the form of 16 bits. The digital ray can be X-ray (X-ray), gamma ray (γ-ray), etc.
[0057] The floating point matrix is denoted as s(x,y), where x and y represent the rows and columns in the matrix respectively.
[0058] Specifically, the matrix elements in the floating-point matrix are restricted in order to prevent the logarithmic function argument from being 0. The restriction value is denoted as s ε , the typical value is 0.01; correspondingly, the floating point limit matrix is recorded as s m (x,y), the calculation formula is as follows:
[0059]
[0060] The grayscale scattering model can be used to approximate the function or effect of the detector when the acquisition accuracy decreases at different levels.
[0061] S2. Grayscale noise model construction:
[0062] The Gaussian blur function is used to transform the 16-bit latent image data of digital rays, and the Gaussian blur function value is restricted according to the requirement that the logarithmic function independent variable in the Gaussian blur function is not 0, and the Gaussian blur restriction matrix, namely the grayscale noise model, is obtained.
[0063] In nature, objects have different appearances depending on the scale of observation. In the image scale space, the blur of each scale image gradually increases, which can simulate the process of the object being formed on the retina when the distance from the object changes from near to far. In short, the larger the scale, the blurrier the image. From the perspective of scale space representation, the Gaussian kernel is the only convolution kernel that can generate multi-scale space. Gaussian blur is the result of processing image data using the Gaussian kernel.
[0064] Gaussian transform has two main functions: one is to use computers to approximate random noise pollution results, and the other is to approximate multi-scale transformation processes. According to relevant optical theories, Gaussian transform is the only transform that can achieve scale transformation results. According to Gaussian transform, random noise pollution results can be approximated and scale transformation results can be achieved.
[0065] Specifically, in this application, the Gaussian blur function is denoted as g(x,y), and its calculation formula is:
[0066] In the formula, δ i Represents the Gaussian blur parameter of the i-th iteration process.
[0067] In the iterative process, the scaling process is approximated by changing different Gaussian blur parameters. According to the formula derivation process, the parameter value of the Gaussian blur parameter is related to the number of iterations. The higher the number of iterations, the finer the fusion result, but the amount of calculation increases sharply. According to the results of multiple experiments, the more reasonable number of iterations is 5. According to the Gaussian parameter calculation formula, the typical values of the Gaussian blur parameter are 0.95, 1.4, 1.85, 2.3, and 2.75.
[0068] Specifically, the Gaussian blur parameter δ i The calculation method is shown in the following formula:
[0069] sigma(ksize)=0.15(ksize-1)+0.5, δ i =sigma(2i+1);
[0070] Where sigma and ksize represent the sign of the Gaussian blur parameter calculation function and the Gaussian kernel size of the i-th iteration.
[0071] Furthermore, in the above Gaussian transformation process, since the Gaussian blur function is essentially a logarithmic function, its function value should also be limited according to the objective requirement that the independent variable of the logarithmic function is not 0. Specifically, the limit value of the Gaussian blur function value is denoted as g ε , and its corresponding Gaussian fuzzy constraint matrix is denoted as g m (x,y), the calculation formula is as follows:
[0072] Furthermore, g ε A typical value of is 0.01.
[0073] The grayscale noise model constructed using the Gaussian fuzzy restriction matrix in the present application can better approximate the noise pollution in the data acquisition and storage process, and approximately obtain the result after the data is polluted by noise.
[0074] S3, adaptive fusion coefficient calculation:
[0075] The indirect main fusion coefficient and the fine-tuning fusion coefficient are calculated using the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating-point restriction matrix.
[0076] First, the indirect fusion constant and the direct fusion constant are also used in the calculation of the indirect main fusion coefficient. In other words, the indirect main fusion coefficient is calculated by combining the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating-point restriction matrix, the indirect fusion constant and the direct fusion constant.
[0077] Furthermore, both the direct fusion constant and the indirect fusion constant are obtained by using actual data and grid search. The typical value of the direct fusion constant α is 0.1, and the typical value of the indirect fusion constant β is 0.2. According to the existing pipeline welding digital ray data, the above typical values have a good effect. The indirect fusion constant and the direct fusion constant can also be called the indirect fusion parameter and the direct fusion parameter.
[0078] Specifically, the indirect main fusion coefficient is denoted as f β (x,y), the calculation formula is:
[0079] In the formula, f α (x,y) is the direct main fusion coefficient, and β is the indirect fusion constant.
[0080] Furthermore, the calculation formula of the direct main fusion coefficient is:
[0081] Where α is the direct fusion constant, l Δ (x, y) is the difference between the logarithmic function values of the corresponding elements in the floating point limit matrix and the Gaussian blur limit matrix.
[0082] Going further, Δ The calculation formula for (x,y) is: Δ (x,y)=l g (x,y)-l s (x,y), where l g (x, y) is the logarithmic function value of the corresponding element in the Gaussian fuzzy restriction matrix, i.e. l g (x,y)=log(g m (x,y)), l s (x, y) is the logarithmic function value of the corresponding element in the floating point limit matrix, i.e. l s (x,y)=log(s m (x,y)).
[0083] Secondly, the fine-tuning constant and the fine-tuning polynomial order are also used in the calculation of the fine-tuning fusion coefficient. In other words, the fine-tuning fusion coefficient is calculated by combining the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating-point restriction matrix, the fine-tuning constant and the fine-tuning polynomial order.
[0084] Further optionally, the fine-tuning constant can also be called a fine-tuning parameter, which is obtained using actual data and network search, and its typical value is 0.618; the order of the fine-tuning polynomial is also obtained using actual data and network search, and its typical value is 5, that is, a 5th-order polynomial, that is, the highest order is 5.
[0085] Specifically, the fine-tuning fusion coefficient is recorded as f γ (x,y), the calculation formula is:
[0086] In the formula, γ is the fine-tuning constant, N is the order of the fine-tuning polynomial, is the jth order of the difference between the logarithmic function values of the corresponding elements in the Gaussian blur limit matrix and the floating-point limit matrix.
[0087] S4. Calculation of logarithmic space fusion results:
[0088] The fusion result of the i-th iteration, i.e., the logarithmic space fusion result, is calculated by using the difference and sum of the logarithmic function values of the corresponding elements in the indirect main fusion coefficient, the fine-tuning fusion coefficient, the Gaussian blur restriction matrix, and the floating-point restriction matrix.
[0089] Specifically, the fusion result of the i-th iteration is Δ i (x,y), the calculation formula is:
[0090] Δ i (x,y)=f β (x,y)l Δ (x,y)+f β (x,y)l Φ (x,y)f γ (x,y); where l Φ (x,y)=l g (x,y)+l s (x,y)
[0091] It is the sum of the logarithmic function values of the corresponding elements in the Gaussian blur limit matrix and the floating-point limit matrix.
[0092] S5. The logarithmic space fusion result is restored to the original space:
[0093] The fusion results of multiple iterations are superimposed, and the normalization method is used to restore the final fusion result after superposition to the original space, realizing the adaptive enhancement of the details of the digital ray 16-bit latent image data.
[0094] Specifically, a typical value of the number of iterations of the multiple iterations is 5 iterations, that is, 5 different scales are used for scale transformation.
[0095] In an embodiment of the present application, a logarithmic function is used to represent light reflection, and a floating-point limit matrix and a Gaussian blur limit matrix of the 16-bit latent image data of the digital ray are used as fusion sources. The logarithmic function is used to calculate the sum and difference of the corresponding element values of the two matrices, and then a variable Gaussian parameter is used to approximate the multi-scale transformation process. The detail enhancement process is implemented through multiple iterations, and the Gaussian parameters are different in each iteration process. The indirect main fusion coefficient and the fine-tuning fusion coefficient corresponding to the 16-bit latent image data of the digital ray are adaptively calculated according to the grayscale noise model and the grayscale scattering model, and the normalization method is used to restore the logarithmic space fusion result to the original space, thereby solving the technical defects of the existing latent image data enhancement technology in processing 16-bit latent image data, such as poor detail enhancement effect, low stability and poor universality.
[0096] Figure 4The hardware architecture diagram of a computer device 10000 suitable for implementing the adaptive enhancement method according to an embodiment of the present application is schematically shown. In some embodiments, the computer device 10000 may be a terminal device such as a smart phone, a wearable device, a tablet computer, a personal computer, a vehicle-mounted terminal, a game console, a virtual device, a workbench, a digital assistant, a set-top box, a robot, etc. In other embodiments, the computer device 10000 may be a rack server, a blade server, a tower server, or a cabinet server (including an independent server, or a server cluster composed of multiple servers), etc. Figure 4 As shown, the computer device 10000 includes but is not limited to: a memory 10010, a processor 10020, and a network interface 10030 which can be interconnected and communicated via a system bus.
[0097] Wherein: the memory 10010 includes at least one type of computer-readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 10010 can be an internal storage module of the computer device 10000, such as a hard disk or memory of the computer device 10000. In other embodiments, the memory 10010 can also be an external storage device of the computer device 10000, such as a plug-in hard disk equipped on the computer device 10000, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 10010 can also include both the internal storage module of the computer device 10000 and its external storage device. In this embodiment, the memory 10010 is generally used to store the operating system and various application software installed in the computer device 10000, such as the program code of the adaptive enhancement method, etc. In addition, the memory 10010 can also be used to temporarily store various data that have been output or will be output.
[0098] In some embodiments, the processor 10020 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other chips. The processor 10020 is generally used to control the overall operation of the computer device 10000, such as performing control and processing related to data interaction or communication with the computer device 10000. In this embodiment, the processor 10020 is used to run the program code stored in the memory 10010 or process data.
[0099] The network interface 10030 may include a wireless network interface or a wired network interface, and the network interface 10030 is generally used to establish a communication link between the computer device 10000 and other computer devices. For example, the network interface 10030 is used to connect the computer device 10000 to an external terminal through a network, and to establish a data transmission channel and a communication link between the computer device 10000 and the external terminal. The network may be a wireless or wired network such as an intranet, the Internet, the Global System of Mobile communication (GSM), Wideband Code Division Multiple Access (WCDMA), 4G network, 5G network, Bluetooth, Wi-Fi, etc.
[0100] It should be pointed out that Figure 4 Only a computer device having components 10010 - 10030 is shown, but it should be understood that implementation of all of the components shown is not a requirement, and more or fewer components may alternatively be implemented.
[0101] In this embodiment, the adaptive enhancement method stored in the memory 10010 can also be divided into one or more program modules and executed by one or more processors (such as processor 10020) to complete the embodiment of the present application.
[0102] A computer-readable storage medium is also provided in an embodiment of the present application. A computer-readable storage medium stores a computer program thereon, wherein when the computer program is executed by a processor, the steps of the adaptive enhancement method in the embodiment are implemented.
[0103] In this embodiment, the computer-readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the computer-readable storage medium can be an internal storage unit of a computer device, such as a hard disk or memory of the computer device. In other embodiments, the computer-readable storage medium can also be an external storage device of a computer device, such as a plug-in hard disk equipped on the computer device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the computer-readable storage medium can also include both the internal storage unit of the computer device and its external storage device. In this embodiment, the computer-readable storage medium is generally used to store an operating system and various application software installed on the computer device, such as the program code of the adaptive enhancement method in the embodiment, etc. In addition, the computer-readable storage medium can also be used to temporarily store various types of data that have been output or are to be output.
[0104] A computer program product is also provided in the embodiment of the present application, including a computer program, which implements the method in the above embodiment when executed by a processor. Obviously, those skilled in the art should understand that the modules or steps of the above-mentioned embodiments of the present application can be implemented with a general-purpose computer device, which can be concentrated on a single computer device, or distributed on a network composed of multiple computer devices, and optionally, they can be implemented with executable program codes of computer devices, so that they can be stored in a storage device and executed by a computer device, and in some cases, the steps shown or described can be performed in a different order from that here, or they can be made into each integrated circuit module respectively, or multiple modules or steps therein can be made into a single integrated circuit module to implement. In this way, the embodiment of the present application is not limited to any specific combination of hardware and software.
[0105] It should be noted that the above are only preferred embodiments of the present application, and the patent protection scope of the present application is not limited thereto. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A method for adaptively enhancing details of 16-bit latent image data of digital rays, characterized in that: include: S1. Grayscale scattering model construction: The 16-bit latent image data of digital rays is mapped to the logarithmic space using a logarithmic function, and the logarithmic function value is restricted according to the requirement that the independent variable of the logarithmic function is not 0, so as to obtain a floating point restriction matrix, namely the grayscale scattering model; S2. Grayscale noise model construction: The 16-bit latent image data of digital rays is transformed by using Gaussian blur function, and the value of Gaussian blur function is restricted according to the requirement that the independent variable of the logarithmic function in the Gaussian blur function is not 0, so as to obtain the Gaussian blur restriction matrix, i.e., the grayscale noise model; S3, adaptive fusion coefficient calculation: The indirect main fusion coefficient and the fine-tuning fusion coefficient are calculated by using the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating point restriction matrix; S4. Calculation of logarithmic space fusion results: The fusion result of the i-th iteration, i.e., the logarithmic space fusion result, is calculated by using the difference and sum of the logarithmic function values of the corresponding elements in the indirect main fusion coefficient, the fine-tuning fusion coefficient, the Gaussian blur restriction matrix, and the floating-point restriction matrix.
2. The adaptive enhancement method according to claim 1, characterized in that: The indirect main fusion coefficient is calculated by combining the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating point restriction matrix, the indirect fusion constant and the direct fusion constant; Among them, the indirect fusion constant and the direct fusion constant are both predetermined based on actual data and grid search.
3. The adaptive enhancement method according to claim 2, characterized in that: The fine-tuning fusion coefficient is calculated by combining the difference between the logarithmic function values of the corresponding elements in the Gaussian blur restriction matrix and the floating-point restriction matrix, the fine-tuning constant and the fine-tuning polynomial order; Among them, the fine-tuning constant and the fine-tuning polynomial order are both predetermined based on actual data and grid search.
4. The adaptive enhancement method according to any one of claims 1 to 3, characterized in that: In S1, the floating point limit matrix is denoted as s m (x,y), the calculation formula is: In the formula, s ε is the limit value of the logarithmic function value, x and y represent the row and column in the matrix respectively; In S2, the Gaussian blur function is denoted as g(x,y), and its calculation formula is: In the formula, δ i Represents the Gaussian blur parameter of the i-th iteration process; sigma(ksize)=0.15(ksize-1)+0.5, δ i =sigma(2i+1), sigma and ksize represent the sign of the Gaussian blur parameter calculation function and the Gaussian kernel size of the i-th iteration; Correspondingly, the Gaussian fuzzy constraint matrix is denoted as g m (x,y), the calculation formula is: In the formula, g ε The limit value of the Gaussian blur function.
5. The adaptive enhancement method according to claim 4, characterized in that: The indirect main fusion coefficient is denoted as f β (x,y), the calculation formula is: In the formula, f α (x, y) is the direct main fusion coefficient, β is the indirect fusion constant; Furthermore, the calculation formula of the direct main fusion coefficient is: Where α is the direct fusion constant, l Δ (x, y) is the difference between the logarithmic function values of the corresponding elements in the floating point limit matrix and the Gaussian blur limit matrix.
6. The adaptive enhancement method according to claim 5, characterized in that: The fine-tuning fusion coefficient is denoted as f γ (x,y), the calculation formula is: In the formula, γ is the fine-tuning constant, N is the order of the fine-tuning polynomial, is the jth order of the difference between the logarithmic function values of the corresponding elements in the Gaussian blur limit matrix and the floating-point limit matrix.
7. The adaptive enhancement method according to claim 6, characterized in that: The fusion result of the i-th iteration is Δ i (x,y), the calculation formula is: Δ i (x,y)=f β (x,y)l Δ (x,y)+f β (x,y)l Φ (x,y)f γ (x,y), or, l Φ (x,y)=l g (x,y)+l s (x,y) It is the sum of the logarithmic function values of the corresponding elements in the Gaussian blur limit matrix and the floating-point limit matrix.
8. A computer device, characterized in that: include: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the adaptive enhancement method according to any one of claims 1 to 7 is implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the adaptive enhancement method according to any one of claims 1 to 7 is implemented.