WO2021175045A1 - 色彩增强方法及相关装置 - Google Patents
色彩增强方法及相关装置 Download PDFInfo
- Publication number
- WO2021175045A1 WO2021175045A1 PCT/CN2021/073598 CN2021073598W WO2021175045A1 WO 2021175045 A1 WO2021175045 A1 WO 2021175045A1 CN 2021073598 W CN2021073598 W CN 2021073598W WO 2021175045 A1 WO2021175045 A1 WO 2021175045A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- color
- resolution
- transformation matrix
- color enhancement
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 92
- 239000011159 matrix material Substances 0.000 claims abstract description 82
- 230000009466 transformation Effects 0.000 claims abstract description 75
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims description 28
- 238000004590 computer program Methods 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 17
- 238000004891 communication Methods 0.000 claims description 13
- 238000013135 deep learning Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 abstract description 14
- 230000009286 beneficial effect Effects 0.000 abstract description 7
- 238000004364 calculation method Methods 0.000 abstract description 6
- 238000004422 calculation algorithm Methods 0.000 abstract description 5
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000013138 pruning Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Definitions
- This application relates to the technical field of electronic equipment, and in particular to a color enhancement method and related devices.
- the color enhancement technology for pictures mainly uses Generative Adversarial Networks (GAN), but the existing GAN is mainly for low-resolution pictures. It will be more difficult to process high-resolution pictures, so it is difficult to process high-resolution pictures. Part of the input resolution is not very large.
- GAN Generative Adversarial Networks
- the embodiments of the present application provide a color enhancement method and related devices, so as to improve the calculation speed and reduce the complexity of the algorithm while ensuring the color enhancement effect.
- an embodiment of the present application provides a color enhancement method, and the method includes:
- the first image to be processed is preprocessed to obtain a second image and a third image.
- the resolution of the second image and the third image are the same and smaller than the first resolution threshold.
- the resolution of the first image is The rate is greater than the second resolution threshold, and the third image is a color-enhanced second image;
- an embodiment of the present application provides a color enhancement device, the color enhancement device includes a processing unit, wherein:
- the processing unit is configured to preprocess the first image to be processed to obtain a second image and a third image, and the second image and the third image have the same resolution and are smaller than a first resolution threshold, The resolution of the first image is greater than the second resolution threshold, the third image is a color-enhanced second image; and the second image and the third image are processed by a least square method.
- the color transformation matrix is obtained by fitting; and used to perform color enhancement on the first image according to the color transformation matrix to obtain a fourth image, and the fourth image is the first image after color enhancement.
- an embodiment of the present application provides an electronic device, including a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be processed by the above
- the above program includes instructions for executing the steps in any method in the first aspect of the embodiments of the present application.
- an embodiment of the present application provides a chip, including: a processor, configured to call and run a computer program from a memory, so that the device installed with the chip executes any method as in the first aspect of the embodiment of the present application Some or all of the steps described in.
- an embodiment of the present application provides a computer-readable storage medium, wherein the above-mentioned computer-readable storage medium stores a computer program for electronic data exchange, wherein the above-mentioned computer program enables a computer to execute On the one hand, part or all of the steps described in any method.
- the embodiments of the present application provide a computer program product, wherein the above-mentioned computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the above-mentioned computer program is operable to cause a computer to execute as implemented in this application.
- the computer program product may be a software installation package.
- FIG. 1 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
- FIG. 2 is a schematic flowchart of a color enhancement method provided by an embodiment of the present application.
- FIG. 3 is a schematic flowchart of another color enhancement method provided by an embodiment of the present application.
- FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
- FIG. 6 is a block diagram of functional units of a color enhancement device provided by an embodiment of the present application.
- the electronic devices involved in the embodiments of this application include electronic devices, which may be electronic devices with communication capabilities, and the electronic devices may include various handheld devices with wireless communication functions, vehicle-mounted devices, wearable devices, and computing devices. Or other processing equipment connected to the wireless modem, as well as various forms of user equipment (User Equipment, UE), mobile station (Mobile Station, MS), terminal equipment (terminal device), and so on.
- UE User Equipment
- MS Mobile Station
- terminal device terminal device
- the electronic device 100 involved in the embodiment of the present application includes a housing 110, a display screen 120, and a motherboard 130.
- the motherboard 130 is provided with a front camera 131, a processor 132, a memory 133, a power management chip 134, etc. .
- the above-mentioned processor 132 is the control center of the electronic device. It uses various interfaces and lines to connect the various parts of the entire electronic device, runs or executes software programs and/or modules stored in the memory 133, and calls Data, perform various functions of electronic equipment and process data, so as to monitor the electronic equipment as a whole.
- the processor 132 may include one or more processing units; preferably, the processor 132 may integrate an application processor and a modem processor, where the application processor mainly processes the operating system, user interface, application programs, etc. , The modem processor mainly deals with wireless communication. It can be understood that the foregoing modem processor may not be integrated into the processor 132.
- the processor 132 may be, for example, a central processing unit (CPU), a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application-Specific Integrated Circuit, ASIC), field programmable Field Programmable Gate Array (FPGA) or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in conjunction with the disclosure of this application.
- the foregoing processor may also be a combination for realizing calculation functions, for example, including a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and so on.
- the aforementioned memory 133 may be used to store software programs and modules.
- the processor 132 executes various functional applications and data processing of the electronic device by running the software programs and modules stored in the memory 133.
- the memory 133 may mainly include a program storage area and a data storage area.
- the program storage area may store an operating system, an application program required by at least one function, and the like; the data storage area may store data created according to the use of an electronic device, etc.
- the memory 133 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.
- the memory 133 can be, for example, random access memory (Random Access Memory, RAM), flash memory, read-only memory (Read Only Memory, ROM), erasable programmable read-only memory (Erasable Programmable ROM, EPROM), and electrically erasable Programmable read-only memory (Electrically EPROM, EEPROM), register, hard disk, mobile hard disk, CD-ROM or any other form of storage medium known in the art.
- RAM Random Access Memory
- ROM read-only memory
- ROM erasable programmable read-only memory
- EPROM erasable programmable read-only memory
- Electrically erasable Programmable read-only memory Electrically erasable Programmable read-only memory
- register hard disk, mobile hard disk, CD-ROM or any other form of storage medium known in the art.
- FIG. 2 is a schematic flowchart of a color enhancement method provided by an embodiment of the present application, and the color enhancement method can be applied to the electronic device shown in FIG. 1. As shown in the figure, the color enhancement method includes the following operations.
- S201 The electronic device preprocesses the first image to be processed to obtain a second image and a third image, where the second image and the third image have the same resolution and are smaller than a first resolution threshold, and the first image The resolution of an image is greater than the second resolution threshold, and the third image is a color-enhanced second image;
- the second image and the third image are images obtained through different preprocessing, and the preprocessing may be, for example, compression processing, cropping processing, and deep learning processing, which is not limited herein.
- the first resolution threshold and the second resolution threshold may be the same or different.
- the first resolution threshold is less than The second resolution threshold.
- the resolution of the second image and the third image may be 100*100, and the resolution of the first image is 1920*1080.
- S202 The electronic device fits the second image and the third image by a least square method to obtain a color transformation matrix
- the electronic device fits the second image and the third image to obtain the color transformation matrix by using the least square method.
- the pixel value of the corresponding pixel in the over-determined equations obtained by fitting the least-squares method, and the over-determined equations corresponds to the color transformation matrix, that is, the color transformation matrix is a non-color-enhanced second image
- S203 The electronic device performs color enhancement on the first image according to the color transformation matrix to obtain a fourth image, where the fourth image is the first image after color enhancement.
- the performing color enhancement of the first image according to the color transformation matrix to obtain a fourth image includes:
- the formed image is the fourth image.
- the electronic device multiplies the first image before color enhancement by the color transformation matrix (that is, the best matching function that minimizes the error between before and after color enhancement), and then the color enhancement can be obtained.
- the fourth image is helpful to reduce the error of color enhancement.
- this solution can be applied to the enhancement of night scene pictures, the beautification of game images, etc., which is not limited here.
- the electronic device preprocesses the first image to be processed to obtain the second image and the third image, and the second image and the third image have the same resolution and are smaller than the first image.
- a resolution threshold the resolution of the first image is greater than the second resolution threshold, and the third image is a second image after color enhancement; the second image and the third image are The image is fitted to obtain a color transformation matrix; the first image is color-enhanced according to the color transformation matrix to obtain a fourth image, and the fourth image is the first image after color enhancement. It can be seen that the electronic device performs color enhancement on the first image to obtain the fourth image through the color transformation matrix obtained by fitting two lower resolution second images and the third image corresponding to the first image to be processed.
- the third image is the second image after color enhancement, which helps to ensure the color enhancement effect of the first image.
- the color transformation can be obtained only through the least square method.
- the matrix does not require deep learning of the first image with a higher resolution, which is beneficial to reduce computational complexity and improve computational efficiency.
- the preprocessing of the first image to be processed to obtain the second image and the third image includes:
- the second image is input into a preset network model to output the third image
- the third image is a second image after color enhancement.
- the preset model is a network model for color enhancement trained through deep learning, for example, it may be a generative model in a generative confrontation network, which is not limited here.
- the preset network model is preset in the electronic device by a technical developer of the electronic device before the electronic device leaves the factory.
- the electronic device obtains a color-enhanced third image by inputting the second image into the deep learning preset model, and inputting the lower-resolution image into the preset model for color enhancement while improving the effect of color enhancement. , To ensure the calculation speed, and provide effect guarantee for the subsequent color enhancement of high-resolution images.
- the preset model is an inference network obtained by training the generative adversarial network GAN through the first reference image and the second reference image, and separating the generative model from the training result, wherein the first reference image
- the reference image and the second reference image include the same image content, and the resolutions of the first reference image and the second reference image are the same, and the resolution of the first reference image and the second reference image Rate is less than the first resolution threshold, and the first reference image is obtained by clipping a third reference image, the second reference image is obtained by clipping a fourth reference image, and the third reference image is The image and the fourth reference image are acquired through different electronic devices.
- the third reference image and the fourth reference image are images obtained by different devices, and one of the two reference images is an image with a higher resolution, and the other is an image with a lower resolution.
- the third reference image is an image with a lower resolution obtained by a smart phone
- the fourth reference image is an image with a higher resolution obtained by a SLR camera
- the third reference image is an image obtained by a SLR camera
- the fourth reference image is an image with a lower resolution obtained through a smart phone, which is not limited here.
- the resolution of the first reference image and the second reference image may be 100*100, 150*150, etc., which are not limited here.
- the preset model is to train the generative adversarial network GAN through the first reference image and the second reference image.
- the generative model is separated and used as the inference network, which is then deployed by the technical developer The electronic equipment.
- the GAN network is trained by cropping a higher resolution image and a lower resolution image to obtain two lower resolution images with the same content, and a convergent generative model is obtained as a preliminary Setting up a model helps to improve the effect of the color enhancement model.
- the method further includes:
- the fourth image is the first image after color enhancement.
- the electronic device detects the fourth image through the peak signal-to-noise ratio to determine whether the color enhancement is successful, and detects the color enhancement effect, which helps ensure the accuracy of color enhancement.
- the method further includes:
- the fourth image is an image after the first image is color-enhanced.
- the electronic device determines the structural similarity between the first image and the fourth image through the structural similarity, which helps to ensure that the fourth image obtained by the color transformation matrix fitted by the least squares method is not distorted. , which helps to ensure the accuracy of color enhancement.
- FIG. 3 is a schematic flowchart of another color enhancement method provided by an embodiment of the present application.
- the color enhancement method can be applied to the electronic device shown in FIG. 1. As shown in the figure, this color enhancement method includes the following operations:
- the electronic device preprocesses the first image to be processed to obtain a second image and a third image, where the second image and the third image have the same resolution and are smaller than a first resolution threshold, and the first image The resolution of an image is greater than the second resolution threshold, and the third image is a color-enhanced second image.
- S302 The electronic device fits the second image and the third image by a least square method to obtain a color transformation matrix.
- the electronic device multiplies the matrix formed by the first pixel value of each pixel in the first image by the color transformation matrix to obtain the second pixel value of each pixel.
- the image formed by the second pixel value is a fourth image.
- S304 The electronic device detects the first image and the fourth image through the structural similarity SSIM.
- the electronic device preprocesses the first image to be processed to obtain the second image and the third image, and the second image and the third image have the same resolution and are smaller than the first image.
- a resolution threshold the resolution of the first image is greater than the second resolution threshold, and the third image is a second image after color enhancement; the second image and the third image are The image is fitted to obtain a color transformation matrix; the first image is color-enhanced according to the color transformation matrix to obtain a fourth image, and the fourth image is the first image after color enhancement. It can be seen that the electronic device performs color enhancement on the first image to obtain the fourth image through the color transformation matrix obtained by fitting two lower resolution second images and the third image corresponding to the first image to be processed.
- the third image is the second image after color enhancement, which helps to ensure the color enhancement effect of the first image.
- the color transformation can be obtained only through the least square method.
- the matrix does not require deep learning of the first image with a higher resolution, which is beneficial to reduce computational complexity and improve computational efficiency.
- the electronic device multiplies the first image before color enhancement by the color transformation matrix (that is, the best matching function that minimizes the error between before and after color enhancement) to obtain the fourth image after color enhancement , which helps reduce the error of color enhancement.
- the color transformation matrix that is, the best matching function that minimizes the error between before and after color enhancement
- the electronic device uses the structural similarity to determine the structural similarity between the first image and the fourth image, which helps to ensure that the fourth image obtained by the color transformation matrix fitted by the least square method of the first image is not distorted, which is beneficial to guarantee Accuracy of color enhancement.
- FIG. 4 is a schematic flowchart of another color enhancement method provided by an embodiment of the present application.
- the color enhancement method can be applied to the electronic device shown in FIG. 1. As shown in the figure, this color enhancement method includes the following operations:
- S401 The electronic device obtains a second image by reducing the first image.
- the electronic device inputs the second image into a preset network model to output a third image, where the third image is a color-enhanced second image, and the resolutions of the second image and the third image are The same and smaller than the first resolution threshold, and the resolution of the first image is greater than the second resolution threshold.
- S403 The electronic device fits the second image and the third image by a least square method to obtain a color transformation matrix.
- S404 The electronic device multiplies the matrix formed by the first pixel value of each pixel in the first image by the color transformation matrix to obtain the second pixel value of each pixel, and the value of each pixel is The image formed by the second pixel value is a fourth image.
- the electronic device determines a peak signal-to-noise ratio PSNR of the fourth image.
- the electronic device preprocesses the first image to be processed to obtain the second image and the third image, and the second image and the third image have the same resolution and are smaller than the first image.
- a resolution threshold the resolution of the first image is greater than the second resolution threshold, and the third image is a second image after color enhancement; the second image and the third image are The image is fitted to obtain a color transformation matrix; the first image is color-enhanced according to the color transformation matrix to obtain a fourth image, and the fourth image is the first image after color enhancement. It can be seen that the electronic device performs color enhancement on the first image to obtain the fourth image through the color transformation matrix obtained by fitting two lower resolution second images and the third image corresponding to the first image to be processed.
- the third image is the second image after color enhancement, which helps to ensure the color enhancement effect of the first image.
- the color transformation can be obtained only through the least square method.
- the matrix does not require deep learning of the first image with a higher resolution, which is beneficial to reduce computational complexity and improve computational efficiency.
- the electronic device inputs the second image into the deep learning preset model to obtain the color-enhanced third image, and inputs the lower-resolution image into the preset model for color enhancement, which improves the effect of color enhancement while ensuring computing Speed, and provide effect guarantee for the subsequent color enhancement of high-resolution images.
- the electronic device multiplies the first image before color enhancement by the color transformation matrix (that is, the best matching function that minimizes the error between before and after color enhancement) to obtain the fourth image after color enhancement , which helps to reduce the error of color enhancement.
- the color transformation matrix that is, the best matching function that minimizes the error between before and after color enhancement
- the electronic device detects the fourth image through the peak signal-to-noise ratio after obtaining the fourth image, determines whether the color enhancement is successful, and detects the color enhancement effect, which helps to ensure the accuracy of the color enhancement.
- FIG. 5 is a schematic structural diagram of an electronic device 500 provided by an embodiment of the present application. 510, memory 520, communication interface 530, and one or more programs 521, where the one or more programs 521 are stored in the memory 520 and are configured to be executed by the application processor 510, and the one or The multiple programs 521 include instructions for performing the following steps:
- the first image to be processed is preprocessed to obtain a second image and a third image.
- the resolution of the second image and the third image are the same and smaller than the first resolution threshold.
- the resolution of the first image is The rate is greater than the second resolution threshold, and the third image is a color-enhanced second image;
- the electronic device preprocesses the first image to be processed to obtain the second image and the third image, and the second image and the third image have the same resolution and are smaller than the first image.
- a resolution threshold the resolution of the first image is greater than the second resolution threshold, and the third image is a second image after color enhancement; the second image and the third image are The image is fitted to obtain a color transformation matrix; the first image is color-enhanced according to the color transformation matrix to obtain a fourth image, and the fourth image is the first image after color enhancement. It can be seen that the electronic device performs color enhancement on the first image to obtain the fourth image through the color transformation matrix obtained by fitting two lower resolution second images and the third image corresponding to the first image to be processed.
- the third image is the second image after color enhancement, which helps to ensure the color enhancement effect of the first image.
- the color transformation can be obtained only through the least square method.
- the matrix does not require deep learning of the first image with a higher resolution, which is beneficial to reduce computational complexity and improve computational efficiency.
- the instructions in the program 521 are specifically used to perform the following operations:
- the image obtains the second image; and is used to input the second image into a preset network model to output the third image, and the third image is a color-enhanced second image.
- the preset model is an inference network obtained by training the generative adversarial network GAN through the first reference image and the second reference image, and separating the generative model from the training result, wherein the first reference image
- the reference image and the second reference image include the same image content, and the resolutions of the first reference image and the second reference image are the same, and the resolution of the first reference image and the second reference image Rate is less than the first resolution threshold, and the first reference image is obtained by clipping a third reference image, the second reference image is obtained by clipping a fourth reference image, and the third reference image is The image and the fourth reference image are acquired through different electronic devices.
- the instructions in the program 521 are specifically used to perform the following operations:
- the matrix composed of the first pixel value of each pixel in the image is multiplied by the color transformation matrix to obtain the second pixel value of each pixel, and the image formed by the second pixel value of each pixel is The fourth image.
- the program 521 further includes instructions for performing the following operations: after the first image is color-enhanced according to the color transformation matrix to obtain a fourth image, the fourth image is determined The peak signal-to-noise ratio PSNR; and when it is detected that the peak signal-to-noise ratio PSNR is greater than a preset threshold, it is determined that the fourth image is the color-enhanced first image.
- the program 521 further includes instructions for performing the following operations: after the first image is color-enhanced according to the color transformation matrix to obtain a fourth image, the SSIM pair The first image and the fourth image are detected; and when the detection is successful, it is determined that the fourth image is an image after the first image is color-enhanced.
- an electronic device includes hardware structures and/or software modules corresponding to each function.
- this application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
- the embodiment of the present application may divide the electronic device into functional units according to the foregoing method examples.
- each functional unit may be divided corresponding to each function, or two or more functions may be integrated into one processing unit.
- the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
- FIG. 6 is a block diagram of the functional unit composition of the color enhancement device 600 involved in an embodiment of the present application.
- the color enhancement device 600 is applied to electronic equipment.
- the color enhancement device includes a processing unit 601, wherein the processing unit 601 is configured to perform any of the steps in the above method embodiments, and the color enhancement device also includes communication The unit 602, when the processing unit 601 performs data transmission such as sending, optionally calls the communication unit 602 to complete corresponding operations.
- the detailed description will be given below.
- the processing unit 601 is configured to preprocess the first image to be processed to obtain a second image and a third image, and the resolution of the second image and the third image are the same and smaller than the first resolution threshold , The resolution of the first image is greater than the second resolution threshold, the third image is a second image after color enhancement; Performing fitting to obtain a color transformation matrix; and for performing color enhancement on the first image according to the color transformation matrix to obtain a fourth image, the fourth image being the first image after color enhancement.
- the electronic device preprocesses the first image to be processed to obtain the second image and the third image, and the second image and the third image have the same resolution and are smaller than the first image.
- a resolution threshold the resolution of the first image is greater than the second resolution threshold, and the third image is a color-enhanced second image; the second image and the third The image is fitted to obtain a color transformation matrix; the first image is color-enhanced according to the color transformation matrix to obtain a fourth image, and the fourth image is the color-enhanced first image.
- the electronic device performs color enhancement on the first image to obtain the fourth image through the color transformation matrix obtained by fitting two lower resolution second images and the third image corresponding to the first image to be processed. Perform operations such as cropping on the first image with higher resolution.
- the third image is the second image after color enhancement, which helps to ensure the color enhancement effect of the first image.
- the color transformation can be obtained only through the least square method. The matrix does not require deep learning of the first image with a higher resolution, which is beneficial to reduce computational complexity and improve computational efficiency.
- the processing unit 601 is specifically configured to: obtain the first image by reducing the first image. Two images; and for inputting the second image into a preset network model through the communication unit 602 to output the third image, the third image being a color-enhanced second image.
- the preset model is an inference network obtained by training the generative adversarial network GAN through the first reference image and the second reference image, and separating the generative model from the training result, wherein the first reference image
- the reference image and the second reference image include the same image content, and the resolutions of the first reference image and the second reference image are the same, and the resolution of the first reference image and the second reference image Rate is less than the first resolution threshold, and the first reference image is obtained by clipping a third reference image, the second reference image is obtained by clipping a fourth reference image, and the third reference image is The image and the fourth reference image are acquired through different electronic devices.
- the processing unit 601 is specifically configured to: The matrix formed by the first pixel values of the dots is multiplied by the color transformation matrix to obtain the second pixel value of each pixel, and the image formed by the second pixel value of each pixel is the fourth image.
- the processing unit 601 after the processing unit 601 performs color enhancement of the first image according to the color transformation matrix to obtain a fourth image, it is further configured to: determine the peak signal-to-noise of the fourth image Ratio PSNR; and when it is detected that the peak signal-to-noise ratio PSNR is greater than a preset threshold, it is determined that the fourth image is the color-enhanced first image.
- the processing unit 601 after the processing unit 601 performs color enhancement of the first image according to the color transformation matrix to obtain a fourth image, it is further configured to: compare the first image with the structural similarity SSIM Performing detection with the fourth image; and when the detection is successful, determining that the fourth image is an image after the first image is color-enhanced.
- the processing unit 601 is specifically configured to: The pixel value of each pixel of the image and the pixel value of the corresponding pixel in the third image are fitted by a least square method to obtain an overdetermined equation set, and the overdetermined equation set is used as the color transformation matrix.
- the color enhancement device 600 may further include a storage unit 603 for storing program codes and data of the electronic device.
- the processing unit 601 may be a processor
- the communication unit 602 may be a touch screen or a transceiver
- the storage unit 603 may be a memory.
- the embodiment of the present application also provides a chip, wherein the chip includes a processor, which is used to call and run a computer program from the memory, so that the device installed with the chip executes the method described in the electronic device in the above method embodiment. Part or all of the steps.
- An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any method as recorded in the above method embodiment ,
- the above-mentioned computer includes electronic equipment.
- the embodiments of the present application also provide a computer program product.
- the above-mentioned computer program product includes a non-transitory computer-readable storage medium storing a computer program. Part or all of the steps of the method.
- the computer program product may be a software installation package, and the above-mentioned computer includes electronic equipment.
- the disclosed device may be implemented in other ways.
- the device embodiments described above are only illustrative, for example, the division of the above-mentioned units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or integrated. To another system, or some features can be ignored, or not implemented.
- the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical or other forms.
- the units described above as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
- the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
- the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
- the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable memory.
- the technical solution of the present application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory.
- a number of instructions are included to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application.
- the aforementioned memory includes: U disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.
- the program can be stored in a computer-readable memory.
- the memory can include: flash disk, Read-Only Memory (ROM), Random Access Memory (RAM), magnetic disk or optical disk, etc.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
本申请实施例公开了一种色彩增强方法及相关装置,方法包括:将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。本申请实施例有利于在保障色彩增强效果的同时提升计算速度,降低算法复杂性。
Description
本申请涉及电子设备技术领域,具体涉及一种色彩增强方法及相关装置。
目前,针对图片的色彩增强技术主要使用生成对抗网络(Generative Adversarial Networks,GAN),但是现有的GAN主要是针对低分辨率的图片,当其处理高分辨率的图片时会比较吃力,所以大部分的输入分辨率都不是很大。
而针对高分辨率图片时,现有的神经网络技术,大多数采用的是卷积神经网络,因为这一类计算适合并行加速。但是,在电子设备(例如智能手机)等终端设备上,部署起来仍然显得比较吃力,其有两种优化方案:其一,采用网络剪枝算法,将网络中不重要的部分取消掉,使其权重稀疏,减小了模型的大小,也减少了算法模型的运行内存,这种优化方式,在GAN的剪枝中会带来较大的精度损失,有的设备甚至不支持稀疏的计算,因此有一定的局限性;其二,采用量化网络模型,使其所有的数据都采用整型8位的数值来表示,这样会将模型优化到原来浮点类型的四分之一,运行内存大大降低,但同样存在着精度损失的问题。
发明内容
本申请实施例提供了一种色彩增强方法及相关装置,以期在保障色彩增强效果的同时提升计算速度,降低算法复杂性。
第一方面,本申请实施例提供一种色彩增强方法,所述方法包括:
将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;
通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;
根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。
第二方面,本申请实施例提供一种色彩增强装置,所述色彩增强装置包括处理单元,其中:
所述处理单元,用于将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;以及用于通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;以及用于根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。
第三方面,本申请实施例提供一种电子设备,包括处理器、存储器、通信接口以及一个或多个程序,其中,上述一个或多个程序被存储在上述存储器中,并且被配置由上述处理器执行,上述程序包括用于执行本申请实施例第一方面任一方法中的步骤的指令。
第四方面,本申请实施例提供了一种芯片,包括:处理器,用于从存储器中调用并运行计算机程序,使得安装有所述芯片的设备执行如本申请实施例第一方面任一方法中所描述的部分或全部步骤。
第五方面,本申请实施例提供了一种计算机可读存储介质,其中,上述计算机可读存储介质存储用于电子数据交换的计算机程序,其中,上述计算机程序使得计算机执行如本申请实施例第一方面任一方法中所描述的部分或全部步骤。
第六方面,本申请实施例提供了一种计算机程序产品,其中,上述计算机程序产品包括存储了计算机程序的非瞬时性计算机可读存储介质,上述计算机程序可操作来使计算机执行如本申请实施例第一方面任一方法中所描述的部分或全部步骤。该计算机程序产品可以为一个软件安装包。
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本申请实施例提供一种电子设备的结构示意图;
图2是本申请实施例提供的一种色彩增强方法的流程示意图;
图3是本申请实施例提供的另一种色彩增强方法的流程示意图;
图4是本申请实施例提供的又一种色彩增强方法的流程示意图;
图5是本申请实施例提供的一种电子设备的结构示意图;
图6是本申请实施例提供的一种色彩增强装置的功能单元组成框图。
为了使本技术领域的人员更好地理解本申请方案,下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别不同对象,而不是用于描述特定顺序。此外,术语“包括”和“具有”以及它们任何变形,意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元,而是可选地还包括没有列出的步骤或单元,或可选地还包括对于这些过程、方法、产品或设备固有的其他步骤或单元。
在本文中提及“实施例”意味着,结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例,也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是,本文所描述的实施例可以与其它实施例相结合。
本申请实施例所涉及到的电子设备包括电子设备,该电子设备可以是具备通信能力的电子设备,该电子设备可以包括各种具有无线通信功能的手持设备、车载设备、可穿戴设备、计算设备或连接到无线调制解调器的其他处理设备,以及各种形式的用户设备(User Equipment,UE),移动台(Mobile Station,MS),终端设备(terminal device)等等。
如图1所示,本申请实施例所涉及的电子设备100包括壳体110、显示屏120、主板130,主板130上设置有前置摄像头131、处理器132、存储器133、电源管理芯片134等。
上述处理器132是电子设备的控制中心,利用各种接口和线路连接整个电子设备的各个部分,通过运行或执行存储在存储器133内的软件程序和/或模块,以及调用存储在存储器133内的数据,执行电子设备的各种功能和处理数据,从而对电子设备进行整体监控。可选的,处理器132可包括一个或多个处理单元;优选的,处理器132可集成应用处理器 和调制解调处理器,其中,应用处理器主要处理操作系统、用户界面和应用程序等,调制解调处理器主要处理无线通信。可以理解的是,上述调制解调处理器也可以不集成到处理器132中。该处理器132例如可以是中央处理器(Central Processing Unit,CPU),通用处理器,数字信号处理器(Digital Signal Processor,DSP),专用集成电路(Application-Specific Integrated Circuit,ASIC),现场可编程门阵列(Field Programmable Gate Array,FPGA)或者其他可编程逻辑器件、晶体管逻辑器件、硬件部件或者其任意组合。其可以实现或执行结合本申请公开内容所描述的各种示例性的逻辑方框,模块和电路。上述处理器也可以是实现计算功能的组合,例如包含一个或多个微处理器组合,DSP和微处理器的组合等等。
上述存储器133可用于存储软件程序以及模块,处理器132通过运行存储在存储器133的软件程序以及模块,从而执行电子设备的各种功能应用以及数据处理。存储器133可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需的应用程序等;存储数据区可存储根据电子设备的使用所创建的数据等。此外,存储器133可以包括高速随机存取存储器,还可以包括非易失性存储器,例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。该存储器133例如可以是随机存取存储器(Random Access Memory,RAM)、闪存、只读存储器(Read Only Memory,ROM)、可擦除可编程只读存储器(Erasable Programmable ROM,EPROM)、电可擦可编程只读存储器(Electrically EPROM,EEPROM)、寄存器、硬盘、移动硬盘、只读光盘(CD-ROM)或者本领域熟知的任何其它形式的存储介质。
请参阅图2,图2是本申请实施例提供了一种色彩增强方法的流程示意图,该色彩增强方法可以应用于如图1所示的电子设备。如图所示,本色彩增强方法包括以下操作。
S201,电子设备将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;
其中,第二图像和第三图像是通过不同的预处理得到的图像,所述预处理例如可以是压缩处理,裁剪处理,深度学习等处理,在此不做限定。
其中,所述第一分辨率阈值和所述第二分辨率阈值可以相同也可以不同,当所述第一分辨率阈值与所述第二分辨率阈值不同时,所述第一分辨率阈值小于所述第二分辨率阈值。
举例而言,所述第二图像和所述第三图像的分辨率可以为100*100,所述第一图像的 分辨率为1920*1080。
S202,所述电子设备通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;
其中,所述电子设备通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵的具体实现方式为将第二图像每个像素点的像素值与第三图像中对应的像素点的像素值通过最小二乘法进行拟合求得的超定方程组,该超定方程组对应的为所述颜色变换矩阵,即该颜色变换矩阵为非色彩增强的第二图像与色彩增强后的第三图像之间的最小化误差的最佳匹配函数。
S203,所述电子设备根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。
在一个可能的示例中,所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,包括:
将所述第一图像中每个像素点的第一像素值组成的矩阵乘以所述颜色变换矩阵得到每个像素点的第二像素值,所述每个像素点的所述第二像素值形成的图像为所述第四图像。
可见,本示例中,电子设备将色彩增强之前的第一图像乘以颜色变换矩阵(即色彩增强之前与色彩增强之后两者之间的最小化误差的最佳匹配函数),可以得到色彩增强之后的第四图像,有利于降低色彩增强的误差。
其中,本方案可以应用于夜景图片的增强、游戏图像的美化等方面,在此不做限定。
可以看出,本申请实施例中,电子设备将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。可见,电子设备通过待处理的第一图像对应的两张分辨率较低的第二图像和第三图像拟合得到的颜色变换矩阵,对第一图像进行色彩增强得到第四图像,不需要对分辨率较高的第一图像进行裁剪等操作,同时,其中的第三图像为色彩增强后的第二图像,有利于保障第一图像色彩增强效果,而且,仅通过最小二乘方法获得颜色变换矩阵,而不需要将分辨率较高的第一图像进行深度学习,有利于降低运算复杂性,提升运算效率。
在一个可能的示例中,所述将待处理的第一图像通过预处理得到第二图像和第三图像, 包括:
通过缩小所述第一图像得到所述第二图像;
将所述第二图像输入预设网络模型输出所述第三图像,所述第三图像为色彩增强后的第二图像。
其中,所述预设模型为通过深度学习训练出的用于色彩增强的网络模型,例如可以是生成对抗网络中的生成模型,在此不做限定。
其中,所述预设网络模型为电子设备的技术开发人员在所述电子设备出厂前预设置在所述电子设备中的。
可见,本示例中,电子设备通过将第二图像输入深度学习类预设模型得到色彩增强后的第三图像,将分辨率较低的图片输入预设模型进行色彩增强在提升色彩增强的效果同时,保障了运算速度,并为后续高分辨率图片进行色彩增强提供了效果保障。
在一个可能的示例中,所述预设模型为通过第一参考图像和第二参考图像对生成对抗网络GAN进行训练,并对训练结果分离出生成模型得到的推理网络,其中,所述第一参考图像与所述第二参考图像包括的图像内容相同,且所述第一参考图像与所述第二参考图像的分辨率相同,且所述第一参考图像和所述第二参考图像的分辨率均小于所述第一分辨率阈值,且所述第一参考图像是通过剪裁第三参考图像得到的,所述第二参考图像是通过剪裁第四参考图像得到的,且所述第三参考图像与所述第四参考图像是通过不同的电子设备获取的。
其中,所述第三参考图像和所述第四参考图像为通过不同的设备获取的图像,且这两张参考图像中一张为分辨率较高的图像,一张为分辨率较低的图像,例如,所述第三参考图像为通过智能手机获取的分辨率较低的图像,第四参考图像为通过单反相机获取的分辨率较高的图像,或者第三参考图像为通过单反相机获取的分辨率较高的图像,第四参考图像为通过智能手机获取的分辨率较低的图像,在此不做限定。
其中,第一参考图像和第二参考图像的分辨率可以是100*100、150*150等,在此不做限定。
其中,所述预设模型为通过第一参考图像和第二参考图像对生成对抗网络GAN进行训练,当训练到收敛状态后,分离出生成模型,作为推理网络,然后,由技术开发人员部署在所述电子设备中。
可见,本示例中,通过裁剪一张分辨率较高的图像和一张分辨率较低的图像得到内容 相同的两张分辨率较低的图像对GAN网络进行训练,得到收敛的生成模型作为预设模型,有利于提升色彩增强模型的效果。
在一个可能的示例中,所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,所述方法还包括:
确定所述第四图像的峰值信噪比PSNR;
当检测到所述峰值信噪比(Peak Signal to Noise Ratio,PSNR)大于预设阈值时,以确定所述第四图像为所述色彩增强后的第一图像。
其中,峰值信噪比越大,图像的色彩越明亮,表明色彩增强成功,因此用以检测第四图像是否为色彩增强后的第一图像。
可见,本示例中,电子设备在得到第四图像后通过峰值信噪比对第四图像进行检测,确定是否色彩增强成功,对色彩增强效果进行检测,有利于保障色彩增强的准确性。
在一个可能的示例中,所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,所述方法还包括:
通过结构相似性(Structural Similarity Index,SSIM)对所述第一图像和所述第四图像进行检测;
当所述检测成功时,确定所述第四图像为所述第一图像进行色彩增强后的图像。
其中,当通过结构相似性检测成功时,表明通过本方案方法得到的色彩增强后的第四图像没有失真,用以保障第四图像为第一图像进行色彩增强后的图像。
可见,本示例中,电子设备通过结构相似性确定第一图像和第四图像之间的结构相似性,有利于确保第一图像通过最小二乘法拟合的颜色变换矩阵得到的第四图像没有失真,有利于保障色彩增强的准确性。
请参阅图3,图3是本申请实施例提供的另一种色彩增强方法的流程示意图,该色彩增强方法可以应用于如图1所示的电子设备。如图所示,本色彩增强方法包括以下操作:
S301,电子设备将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像。
S302,所述电子设备通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵。
S303,所述电子设备将所述第一图像中每个像素点的第一像素值组成的矩阵乘以所述颜色变换矩阵得到每个像素点的第二像素值,所述每个像素点的所述第二像素值形成的图像为第四图像。
S304,所述电子设备通过结构相似性SSIM对所述第一图像和所述第四图像进行检测。
S305,所述电子设备当所述检测成功时,确定所述第四图像为所述第一图像进行色彩增强后的图像。
可以看出,本申请实施例中,电子设备将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。可见,电子设备通过待处理的第一图像对应的两张分辨率较低的第二图像和第三图像拟合得到的颜色变换矩阵,对第一图像进行色彩增强得到第四图像,不需要对分辨率较高的第一图像进行裁剪等操作,同时,其中的第三图像为色彩增强后的第二图像,有利于保障第一图像色彩增强效果,而且,仅通过最小二乘方法获得颜色变换矩阵,而不需要将分辨率较高的第一图像进行深度学习,有利于降低运算复杂性,提升运算效率。
此外,电子设备将色彩增强之前的第一图像乘以颜色变换矩阵(即色彩增强之前与色彩增强之后两者之间的最小化误差的最佳匹配函数),可以得到色彩增强之后的第四图像,有利于降低色彩增强的误差。
此外,电子设备通过结构相似性确定第一图像和第四图像之间的结构相似性,有利于确保第一图像通过最小二乘法拟合的颜色变换矩阵得到的第四图像没有失真,有利于保障色彩增强的准确性。
请参阅图4,图4是本申请实施例提供的另一种色彩增强方法的流程示意图,该色彩增强方法可以应用于如图1所示的电子设备。如图所示,本色彩增强方法包括以下操作:
S401,电子设备通过缩小第一图像得到第二图像。
S402,所述电子设备将所述第二图像输入预设网络模型输出第三图像,所述第三图像为色彩增强后的第二图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值。
S403,所述电子设备通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵。
S404,所述电子设备将所述第一图像中每个像素点的第一像素值组成的矩阵乘以所述颜色变换矩阵得到每个像素点的第二像素值,所述每个像素点的所述第二像素值形成的图像为第四图像。
S405,所述电子设备确定所述第四图像的峰值信噪比PSNR。
S406,所述电子设备当检测到所述峰值信噪比PSNR大于预设阈值时,以确定所述第四图像为色彩增强后的第一图像。
可以看出,本申请实施例中,电子设备将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。可见,电子设备通过待处理的第一图像对应的两张分辨率较低的第二图像和第三图像拟合得到的颜色变换矩阵,对第一图像进行色彩增强得到第四图像,不需要对分辨率较高的第一图像进行裁剪等操作,同时,其中的第三图像为色彩增强后的第二图像,有利于保障第一图像色彩增强效果,而且,仅通过最小二乘方法获得颜色变换矩阵,而不需要将分辨率较高的第一图像进行深度学习,有利于降低运算复杂性,提升运算效率。
此外,电子设备通过将第二图像输入深度学习类预设模型得到色彩增强后的第三图像,将分辨率较低的图片输入预设模型进行色彩增强在提升色彩增强的效果同时,保障了运算速度,并为后续高分辨率图片进行色彩增强提供了效果保障。
此外,电子设备将色彩增强之前的第一图像乘以颜色变换矩阵(即色彩增强之前与色彩增强之后两者之间的最小化误差的最佳匹配函数),可以得到色彩增强之后的第四图像,有利于降低色彩增强的误差。
此外,电子设备在得到第四图像后通过峰值信噪比对第四图像进行检测,确定是否色彩增强成功,对色彩增强效果进行检测,有利于保障色彩增强的准确性。
与上述图2、图3、图4所示的实施例一致的,请参阅图5,图5是本申请实施例提供的一种电子设备500的结构示意图,所述电子设备500还包括应用处理器510、存储器520、 通信接口530以及一个或多个程序521,其中,所述一个或多个程序521被存储在上述存储器520中,并且被配置由上述应用处理器510执行,所述一个或多个程序521包括用于执行以下步骤的指令:
将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;
通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;
根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。
可以看出,本申请实施例中,电子设备将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。可见,电子设备通过待处理的第一图像对应的两张分辨率较低的第二图像和第三图像拟合得到的颜色变换矩阵,对第一图像进行色彩增强得到第四图像,不需要对分辨率较高的第一图像进行裁剪等操作,同时,其中的第三图像为色彩增强后的第二图像,有利于保障第一图像色彩增强效果,而且,仅通过最小二乘方法获得颜色变换矩阵,而不需要将分辨率较高的第一图像进行深度学习,有利于降低运算复杂性,提升运算效率。
在一个可能的示例中,在所述将待处理的第一图像通过预处理得到第二图像和第三图像方面,所述程序521中的指令具体用于执行以下操作:通过缩小所述第一图像得到所述第二图像;以及用于将所述第二图像输入预设网络模型输出所述第三图像,所述第三图像为色彩增强后的第二图像。
在一个可能的示例中,所述预设模型为通过第一参考图像和第二参考图像对生成对抗网络GAN进行训练,并对训练结果分离出生成模型得到的推理网络,其中,所述第一参考图像与所述第二参考图像包括的图像内容相同,且所述第一参考图像与所述第二参考图像的分辨率相同,且所述第一参考图像和所述第二参考图像的分辨率均小于所述第一分辨率阈值,且所述第一参考图像是通过剪裁第三参考图像得到的,所述第二参考图像是通过剪裁第四参考图像得到的,且所述第三参考图像与所述第四参考图像是通过不同的电子设备 获取的。
在一个可能的示例中,在所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像方面,所述程序521中的指令具体用于执行以下操作:将所述第一图像中每个像素点的第一像素值组成的矩阵乘以所述颜色变换矩阵得到每个像素点的第二像素值,所述每个像素点的所述第二像素值形成的图像为所述第四图像。
在一个可能的示例中,所述程序521还包括用于执行以下操作的指令:所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,确定所述第四图像的峰值信噪比PSNR;以及当检测到所述峰值信噪比PSNR大于预设阈值时,以确定所述第四图像为所述色彩增强后的第一图像。
在一个可能的示例中,所述程序521还包括用于执行以下操作的指令:所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,通过结构相似性SSIM对所述第一图像和所述第四图像进行检测;以及当所述检测成功时,确定所述第四图像为所述第一图像进行色彩增强后的图像。
上述主要从方法侧执行过程的角度对本申请实施例的方案进行了介绍。可以理解的是,电子设备为了实现上述功能,其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到,结合本文中所提供的实施例描述的各示例的单元及算法步骤,本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对电子设备进行功能单元的划分,例如,可以对应各个功能划分各个功能单元,也可以将两个或两个以上的功能集成在一个处理单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。需要说明的是,本申请实施例中对单元的划分是示意性的,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式。
图6是本申请实施例中所涉及的色彩增强装置600的功能单元组成框图。该色彩增强装置600应用于电子设备,所述色彩增强装置包括处理单元601,其中,所述处理单元601,用于执行如上述方法实施例中的任一步骤,所述色彩增强装置还包括通信单元602,在所 述处理单元601执行诸如发送等数据传输时,可选择的调用所述通信单元602来完成相应操作。下面进行详细说明。
所述处理单元601,用于将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;以及用于通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;以及用于根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。
可以看出,本申请实施例中,电子设备将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。可见,电子设备通过待处理的第一图像对应的两张分辨率较低的第二图像和第三图像拟合得到的颜色变换矩阵,对第一图像进行色彩增强得到第四图像,不需要对分辨率较高的第一图像进行裁剪等操作,同时,其中的第三图像为色彩增强后的第二图像,有利于保障第一图像色彩增强效果,而且,仅通过最小二乘方法获得颜色变换矩阵,而不需要将分辨率较高的第一图像进行深度学习,有利于降低运算复杂性,提升运算效率。
在一个可能的示例中,在所述将待处理的第一图像通过预处理得到第二图像和第三图像方面,所述处理单元601具体用于:通过缩小所述第一图像得到所述第二图像;以及用于通过通信单元602将所述第二图像输入预设网络模型输出所述第三图像,所述第三图像为色彩增强后的第二图像。
在一个可能的示例中,所述预设模型为通过第一参考图像和第二参考图像对生成对抗网络GAN进行训练,并对训练结果分离出生成模型得到的推理网络,其中,所述第一参考图像与所述第二参考图像包括的图像内容相同,且所述第一参考图像与所述第二参考图像的分辨率相同,且所述第一参考图像和所述第二参考图像的分辨率均小于所述第一分辨率阈值,且所述第一参考图像是通过剪裁第三参考图像得到的,所述第二参考图像是通过剪裁第四参考图像得到的,且所述第三参考图像与所述第四参考图像是通过不同的电子设备获取的。
在一个可能的示例中,在所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像方面,所述处理单元601具体用于:将所述第一图像中每个像素点的第一像素值组成的矩阵乘以所述颜色变换矩阵得到每个像素点的第二像素值,所述每个像素点的所述第二像素值形成的图像为所述第四图像。
在一个可能的示例中,所述处理单元601在所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,还用于:确定所述第四图像的峰值信噪比PSNR;以及当检测到所述峰值信噪比PSNR大于预设阈值时,以确定所述第四图像为所述色彩增强后的第一图像。
在一个可能的示例中,所述处理单元601所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,还用于:通过结构相似性SSIM对所述第一图像和所述第四图像进行检测;以及当所述检测成功时,确定所述第四图像为所述第一图像进行色彩增强后的图像。
在一个可能的示例中,在所述通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵方面,所述处理单元601具体用于:将所述第二图像每个像素点的像素值与所述第三图像中对应的像素点的像素值通过最小二乘法进行拟合求得超定方程组,将该超定方程组作为所述颜色变换矩阵。
其中,所述色彩增强装置600还可以包括存储单元603,用于存储电子设备的程序代码和数据。所述处理单元601可以是处理器,所述通信单元602可以是触控显示屏或者收发器,存储单元603可以是存储器。
可以理解的是,由于方法实施例与装置实施例为相同技术构思的不同呈现形式,因此,本申请中方法实施例部分的内容应同步适配于装置实施例部分,此处不再赘述。
本申请实施例还提供了一种芯片,其中,该芯片包括处理器,用于从存储器中调用并运行计算机程序,使得安装有所述芯片的设备执行如上述方法实施例中电子设备所描述的部分或全部步骤。
本申请实施例还提供一种计算机存储介质,其中,该计算机存储介质存储用于电子数据交换的计算机程序,该计算机程序使得计算机执行如上述方法实施例中记载的任一方法的部分或全部步骤,上述计算机包括电子设备。
本申请实施例还提供一种计算机程序产品,上述计算机程序产品包括存储了计算机程 序的非瞬时性计算机可读存储介质,上述计算机程序可操作来使计算机执行如上述方法实施例中记载的任一方法的部分或全部步骤。该计算机程序产品可以为一个软件安装包,上述计算机包括电子设备。
需要说明的是,对于前述的各方法实施例,为了简单描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本申请并不受所描述的动作顺序的限制,因为依据本申请,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作和模块并不一定是本申请所必须的。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述的部分,可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中,应该理解到,所揭露的装置,可通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如上述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性或其它的形式。
上述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
上述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储器中。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储器中,包括若干指令用以使得一台计算机设备(可为个人计算机、服务器或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储器包括:U盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的 介质。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成,该程序可以存储于计算机可读存储器中,存储器可以包括:闪存盘、只读存储器(Read-Only Memory,ROM)、随机存取器(Random Access Memory,RAM)、磁盘或光盘等。
以上对本申请实施例进行了详细介绍,本文中应用了具体个例对本申请的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本申请的方法及其核心思想;同时,对于本领域的一般技术人员,依据本申请的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本申请的限制。
Claims (20)
- 一种色彩增强方法,其特征在于,所述方法包括:将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。
- 根据权利要求1所述的方法,其特征在于,所述将待处理的第一图像通过预处理得到第二图像和第三图像,包括:通过缩小所述第一图像得到所述第二图像;将所述第二图像输入预设网络模型输出所述第三图像,所述第三图像为色彩增强后的第二图像。
- 根据权利要求2所述的方法,其特征在于,所述预设网络模型为通过深度学习训练出的用于色彩增强的网络模型。
- 根据权利要求2所述的方法,其特征在于,所述预设模型为通过第一参考图像和第二参考图像对生成对抗网络GAN进行训练,并对训练结果分离出生成模型得到的推理网络,其中,所述第一参考图像与所述第二参考图像包括的图像内容相同,且所述第一参考图像与所述第二参考图像的分辨率相同,且所述第一参考图像和所述第二参考图像的分辨率均小于所述第一分辨率阈值,且所述第一参考图像是通过剪裁第三参考图像得到的,所述第二参考图像是通过剪裁第四参考图像得到的,且所述第三参考图像与所述第四参考图像是通过不同的电子设备获取的。
- 根据权利要求1-4任一项所述的方法,其特征在于,所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,包括:将所述第一图像中每个像素点的第一像素值组成的矩阵乘以所述颜色变换矩阵得到每个像素点的第二像素值,所述每个像素点的所述第二像素值形成的图像为所述第四图像。
- 根据权利要求5所述的方法,其特征在于,所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,所述方法还包括:确定所述第四图像的峰值信噪比PSNR;当检测到所述峰值信噪比PSNR大于预设阈值时,以确定所述第四图像为所述色彩增强后的第一图像。
- 根据权利要求1-4任一项所述的方法,其特征在于,所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,所述方法还包括:通过结构相似性SSIM对所述第一图像和所述第四图像进行检测;当所述检测成功时,确定所述第四图像为所述第一图像进行色彩增强后的图像。
- 根据权利要求1所述的方法,其特征在于,所述通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵,包括:将所述第二图像每个像素点的像素值与所述第三图像中对应的像素点的像素值通过最小二乘法进行拟合求得超定方程组,将该超定方程组作为所述颜色变换矩阵。
- 根据权利要求8所述的方法,其特征在于,所述颜色变换矩阵为非色彩增强的所述第二图像与色彩增强后的所述第三图像之间的最小化误差的最佳匹配函数。
- 一种色彩增强装置,其特征在于,所述色彩增强装置包括处理单元,其中:所述处理单元,用于将待处理的第一图像通过预处理得到第二图像和第三图像,所述第二图像和所述第三图像的分辨率相同,且小于第一分辨率阈值,所述第一图像的分辨率大于第二分辨率阈值,所述第三图像为色彩增强后的第二图像;以及用于通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵;以及用于根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像,所述第四图像为色彩增强后的第一图像。
- 根据权利要求10所述的装置,其特征在于,在所述将待处理的第一图像通过预处理得到第二图像和第三图像方面,所述处理单元具体用于:通过缩小所述第一图像得到所述第二图像;以及用于通过通信单元将所述第二图像输入预设网络模型输出所述第三图像,所述第三图像为色彩增强后的第二图像。
- 根据权利要求11所述的装置,其特征在于,所述预设网络模型为通过深度学习训练出的用于色彩增强的网络模型。
- 根据权利要求11所述的装置,其特征在于,所述预设模型为通过第一参考图像和第二参考图像对生成对抗网络GAN进行训练,并对训练结果分离出生成模型得到的推理网络,其中,所述第一参考图像与所述第二参考图像包括的图像内容相同,且所述第一参考 图像与所述第二参考图像的分辨率相同,且所述第一参考图像和所述第二参考图像的分辨率均小于所述第一分辨率阈值,且所述第一参考图像是通过剪裁第三参考图像得到的,所述第二参考图像是通过剪裁第四参考图像得到的,且所述第三参考图像与所述第四参考图像是通过不同的电子设备获取的。
- 根据权利要求10-13任一项所述的装置,其特征在于,在所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像方面,所述处理单元具体用于:将所述第一图像中每个像素点的第一像素值组成的矩阵乘以所述颜色变换矩阵得到每个像素点的第二像素值,所述每个像素点的所述第二像素值形成的图像为所述第四图像。
- 根据权利要求10所述的装置,其特征在于,所述处理单元在所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,还用于:确定所述第四图像的峰值信噪比PSNR;以及当检测到所述峰值信噪比PSNR大于预设阈值时,以确定所述第四图像为所述色彩增强后的第一图像。
- 根据权利要求10-13任一项所述的装置,其特征在于,所述处理单元所述根据所述颜色变换矩阵将所述第一图像进行色彩增强得到第四图像之后,还用于:通过结构相似性SSIM对所述第一图像和所述第四图像进行检测;以及当所述检测成功时,确定所述第四图像为所述第一图像进行色彩增强后的图像。
- 根据权利要求10所述的装置,其特征在于,在所述通过最小二乘方法将所述第二图像和所述第三图像进行拟合得到颜色变换矩阵方面,所述处理单元具体用于:将所述第二图像每个像素点的像素值与所述第三图像中对应的像素点的像素值通过最小二乘法进行拟合求得超定方程组,将该超定方程组作为所述颜色变换矩阵。
- 一种芯片,其特征在于,包括:处理器,用于从存储器中调用并运行计算机程序,使得安装有所述芯片的设备执行如权利要求1-9中任一项所述的方法。
- 一种电子设备,其特征在于,包括处理器、存储器、通信接口,以及一个或多个程序,所述一个或多个程序被存储在所述存储器中,并且被配置由所述处理器执行,所述程序包括用于执行如权利要求1-9任一项所述的方法中的步骤的指令。
- 一种计算机可读存储介质,其特征在于,存储用于电子数据交换的计算机程序,其中,所述计算机程序使得计算机执行如权利要求1-9任一项所述的方法。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010137897.8A CN111145128B (zh) | 2020-03-02 | 2020-03-02 | 色彩增强方法及相关装置 |
CN202010137897.8 | 2020-03-02 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021175045A1 true WO2021175045A1 (zh) | 2021-09-10 |
Family
ID=70528252
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2021/073598 WO2021175045A1 (zh) | 2020-03-02 | 2021-01-25 | 色彩增强方法及相关装置 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111145128B (zh) |
WO (1) | WO2021175045A1 (zh) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111145128B (zh) * | 2020-03-02 | 2023-05-26 | Oppo广东移动通信有限公司 | 色彩增强方法及相关装置 |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1992780A (zh) * | 2005-12-27 | 2007-07-04 | 华晶科技股份有限公司 | 产生较低分辨率的图像的色彩信息重建方法 |
CN102110289A (zh) * | 2011-03-29 | 2011-06-29 | 东南大学 | 基于变分框架的彩色图像对比度增强方法 |
CN109559276A (zh) * | 2018-11-14 | 2019-04-02 | 武汉大学 | 一种基于无参考质量评价和特征统计的图像超分辨率重建方法 |
CN110197471A (zh) * | 2019-03-01 | 2019-09-03 | 合肥工业大学 | 一种图像对比度增强方法 |
CN110363720A (zh) * | 2019-07-04 | 2019-10-22 | 北京奇艺世纪科技有限公司 | 一种图像的色彩增强方法、装置、设备和存储介质 |
US20190333199A1 (en) * | 2018-04-26 | 2019-10-31 | The Regents Of The University Of California | Systems and methods for deep learning microscopy |
CN110634170A (zh) * | 2019-08-30 | 2019-12-31 | 福建帝视信息科技有限公司 | 一种基于语义内容和快速图像检索的照片级图像生成方法 |
CN111145128A (zh) * | 2020-03-02 | 2020-05-12 | Oppo广东移动通信有限公司 | 色彩增强方法及相关装置 |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3409541B2 (ja) * | 1995-11-14 | 2003-05-26 | 三菱電機株式会社 | 色補正方法及び色補正装置並びに色補正応用装置及びカラー画像システム |
JP2005079834A (ja) * | 2003-08-29 | 2005-03-24 | Pentax Corp | 色変換マトリクス算出方法および画像信号処理装置 |
JP5552795B2 (ja) * | 2009-10-26 | 2014-07-16 | 株式会社ニコン | 撮像装置、画像処理装置およびプログラム |
CN104935902B (zh) * | 2015-06-02 | 2017-03-15 | 三星电子(中国)研发中心 | 图像色彩增强方法、装置及电子设备 |
CN109087269B (zh) * | 2018-08-21 | 2020-08-04 | 厦门美图之家科技有限公司 | 弱光图像增强方法及装置 |
CN109903250B (zh) * | 2019-02-25 | 2022-10-04 | 大连海事大学 | 基于多尺度梯度域对比度拉伸的水下图像清晰化处理方法 |
CN110120024B (zh) * | 2019-05-20 | 2021-08-17 | 百度在线网络技术(北京)有限公司 | 图像处理的方法、装置、设备和存储介质 |
-
2020
- 2020-03-02 CN CN202010137897.8A patent/CN111145128B/zh active Active
-
2021
- 2021-01-25 WO PCT/CN2021/073598 patent/WO2021175045A1/zh active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1992780A (zh) * | 2005-12-27 | 2007-07-04 | 华晶科技股份有限公司 | 产生较低分辨率的图像的色彩信息重建方法 |
CN102110289A (zh) * | 2011-03-29 | 2011-06-29 | 东南大学 | 基于变分框架的彩色图像对比度增强方法 |
US20190333199A1 (en) * | 2018-04-26 | 2019-10-31 | The Regents Of The University Of California | Systems and methods for deep learning microscopy |
CN109559276A (zh) * | 2018-11-14 | 2019-04-02 | 武汉大学 | 一种基于无参考质量评价和特征统计的图像超分辨率重建方法 |
CN110197471A (zh) * | 2019-03-01 | 2019-09-03 | 合肥工业大学 | 一种图像对比度增强方法 |
CN110363720A (zh) * | 2019-07-04 | 2019-10-22 | 北京奇艺世纪科技有限公司 | 一种图像的色彩增强方法、装置、设备和存储介质 |
CN110634170A (zh) * | 2019-08-30 | 2019-12-31 | 福建帝视信息科技有限公司 | 一种基于语义内容和快速图像检索的照片级图像生成方法 |
CN111145128A (zh) * | 2020-03-02 | 2020-05-12 | Oppo广东移动通信有限公司 | 色彩增强方法及相关装置 |
Also Published As
Publication number | Publication date |
---|---|
CN111145128B (zh) | 2023-05-26 |
CN111145128A (zh) | 2020-05-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110263909B (zh) | 图像识别方法及装置 | |
WO2021184972A1 (zh) | 图像分割方法、装置、电子设备及存储介质 | |
US20210272306A1 (en) | Method for training image depth estimation model and method for processing image depth information | |
CN112308806B (zh) | 图像处理方法、装置、电子设备以及可读存储介质 | |
CN110929865B (zh) | 网络量化方法、业务处理方法及相关产品 | |
CN107909583B (zh) | 一种图像处理方法、装置及终端 | |
WO2017096946A1 (zh) | 图像高频信息的定位方法和装置 | |
KR101620933B1 (ko) | 제스쳐 인식 메커니즘을 제공하는 방법 및 장치 | |
WO2018045789A1 (zh) | 图像灰度值调整方法和装置 | |
CN114092963B (zh) | 关键点检测及模型训练方法、装置、设备和存储介质 | |
CN112614110B (zh) | 评估图像质量的方法、装置及终端设备 | |
TWI808177B (zh) | 用於連續多幀圖像去除雜訊的裝置和方法 | |
WO2020097836A1 (zh) | 图像处理方法、装置、计算机设备和存储介质 | |
WO2021175040A1 (zh) | 视频处理方法及相关装置 | |
US10210630B2 (en) | Picture compression method and apparatus, and mobile terminal | |
WO2020134789A1 (zh) | 移动终端及其屏幕亮灭的控制方法、计算机存储介质 | |
WO2021175045A1 (zh) | 色彩增强方法及相关装置 | |
US20230005171A1 (en) | Visual positioning method, related apparatus and computer program product | |
US11238563B2 (en) | Noise processing method and apparatus | |
WO2022121701A1 (zh) | 图像处理方法、装置、电子设备以及存储介质 | |
WO2020134547A1 (zh) | 数据的定点化加速方法、装置、电子设备及存储介质 | |
WO2024022060A1 (zh) | 一种图像配准方法、装置及存储介质 | |
CN112862095A (zh) | 基于特征分析的自蒸馏学习方法、设备以及可读存储介质 | |
CN116756444A (zh) | 图像处理方法、装置、设备以及存储介质 | |
CN111489289A (zh) | 一种图像处理方法、图像处理装置及终端设备 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21765158 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21765158 Country of ref document: EP Kind code of ref document: A1 |