WO2021036442A1 - Procédé et appareil de filtration lisse préservant les bords cycliques, et dispositif électronique - Google Patents

Procédé et appareil de filtration lisse préservant les bords cycliques, et dispositif électronique Download PDF

Info

Publication number
WO2021036442A1
WO2021036442A1 PCT/CN2020/097641 CN2020097641W WO2021036442A1 WO 2021036442 A1 WO2021036442 A1 WO 2021036442A1 CN 2020097641 W CN2020097641 W CN 2020097641W WO 2021036442 A1 WO2021036442 A1 WO 2021036442A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
filtered
edge
filtering
preserving smoothing
Prior art date
Application number
PCT/CN2020/097641
Other languages
English (en)
Chinese (zh)
Inventor
聂旎
张哲斌
Original Assignee
北京迈格威科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京迈格威科技有限公司 filed Critical 北京迈格威科技有限公司
Publication of WO2021036442A1 publication Critical patent/WO2021036442A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing

Definitions

  • the present invention relates to the technical field of image processing, in particular to a method, device and electronic equipment for cyclic edge preserving smooth filtering.
  • Edge-preserving smoothing filtering is a filtering algorithm widely used in computer vision, computer graphics and other tasks. Designed to smooth the details of the image while preserving the edges in the image.
  • the algorithms used in the industry are mostly circular edge-preserving smoothing filters (for example: domain transform filter), which are better than other types of edge-preserving smoothing filters based on bilateral filters, the latter It usually takes a long time, and the accelerated version is mostly realized based on the idea of quantization, and the final image effect is poor.
  • the loop edge-preserving smoothing filter ensures that the ideal filtering effect is achieved in all scenes of the image to be filtered.
  • the number of loop iterations of the loop-preserving smoothing filter is usually a very large fixed value.
  • the filtering effect of the loop edge preserving smoothing filter is related to the scene of the image to be filtered. For some scenes of the image to be filtered, when the loop edge preserving smoothing filter is performed, there is no need to perform the above-mentioned multiple loop iterations at all. It can also achieve a considerable filtering effect.
  • the existing loop edge-preserving smoothing filter takes a long time for image processing due to the large number of loop iterations.
  • embodiments of the present invention are proposed to provide a method, device and electronic device for cyclic edge preserving smoothing filtering that overcomes the above-mentioned problems or at least partially solves the above-mentioned problems.
  • the first aspect of the present invention provides a method of cyclic edge preserving smoothing and filtering, including:
  • a device for cyclic edge preserving smoothing and filtering including:
  • An acquiring unit configured to acquire a color image of the target scene and an initial image to be filtered corresponding to the color image
  • An initialization unit configured to use the initial image to be filtered as the first image to be filtered
  • the loop edge preserving smooth filter processing unit is used to perform the following loop edge preserving smooth filter processing:
  • an electronic device including:
  • a computer storage medium having a computer program stored thereon, and the computer executes the above-mentioned cyclic edge-preserving smoothing filtering method when the computer runs the computer program.
  • the color image of the target scene and the initial image to be filtered corresponding to the color image are first acquired; then, the initial image to be filtered is taken as the first image to be filtered; the following cyclic edge-preserving smoothing filtering process is performed:
  • the color image performs edge-preserving smoothing filtering processing on the current image to be filtered to obtain the current smoothed filtered image; further, based on the current smoothed filtered image and the initial image to be filtered, the current image to be filtered is updated, and based on the updated image to be filtered Determine whether to perform the next cycle edge-preserving smoothing filtering process with this smoothing and filtering image and the initial image to be filtered.
  • this method will determine whether to perform the next loop edge-preserving smoothing filter based on the updated image to be filtered and the current smooth-filtered image, as well as the initial image to be filtered, each time the loop edge-preserving smoothing filtering process is performed.
  • the method of cyclic edge-preserving smoothing filtering can determine whether the cyclic edge-preserving smoothing filtering process can be ended in advance based on the updated image to be filtered and the current smoothed filtered image, as well as the initial image to be filtered, without reaching the preset
  • the number of cycles that is, the method can adapt to the scene of the color image, reduces the time consumption of image processing, and alleviates the technical problem that the existing loop edge-preserving smoothing filter takes a long time in image processing.
  • FIG. 1 is a schematic diagram of an electronic device provided by an embodiment of the present invention.
  • FIG. 2 is a flowchart of a method for cyclic edge preserving smoothing filtering provided by an embodiment of the present invention
  • FIG. 3 is a flowchart of another method for cyclic edge preserving smoothing filtering provided by an embodiment of the present invention.
  • FIG. 4 is a flowchart of a method for determining whether to perform the next loop edge-preserving smoothing filtering process based on the updated image to be filtered, the current smoothed filtered image, and the initial image to be filtered according to an embodiment of the present invention
  • FIG. 5 is a flowchart of a method for determining whether to perform the next loop edge-preserving smoothing filtering process according to the first similarity degree and the second similarity degree according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of a device for cyclic edge preserving smoothing and filtering provided by an embodiment of the present invention
  • Fig. 7 schematically shows a block diagram of a computing processing device for executing the method according to the present invention
  • Fig. 8 schematically shows a storage unit for holding or carrying program codes for implementing the method according to the present invention.
  • the electronic device 100 for implementing an embodiment of the present invention will be described.
  • the electronic device can be used to run the cyclic edge-preserving smoothing filtering method of each embodiment of the present invention.
  • the electronic device 100 includes one or more processors 102, one or more memories 104, an input device 106, an output device 108, and a camera 110. These components pass through a bus system 112 and/or other forms of connection mechanisms. (Not shown) interconnected. It should be noted that the components and structure of the electronic device 100 shown in FIG. 1 are only exemplary and not restrictive, and the electronic device may also have other components and structures as required.
  • the processor 102 may adopt a digital signal processor (DSP, Digital Signal Processing), a field programmable gate array (FPGA, Field-Programmable Gate Array), a programmable logic array (PLA, Programmable Logic Array), and an ASIC (Application Specific).
  • DSP Digital Signal Processing
  • FPGA field programmable gate array
  • PLA programmable logic array
  • ASIC Application Specific
  • the processor 102 may be a central processing unit (CPU, Central Processing Unit) or other forms of processing units with data processing capabilities and/or instruction execution capabilities, and may Control other components in the electronic device 100 to perform desired functions.
  • the memory 104 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory.
  • the volatile memory may include random access memory (RAM) and/or cache memory (cache), for example.
  • the non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, and the like.
  • One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 102 may run the program instructions to implement the client functions (implemented by the processor) in the embodiments of the present invention described below. And/or other desired functions.
  • Various application programs and various data such as various data used and/or generated by the application program, can also be stored in the computer-readable storage medium.
  • the input device 106 may be a device used by the user to input instructions, and may include one or more of a keyboard, a mouse, a microphone, and a touch screen.
  • the output device 108 may output various information (for example, images or sounds) to the outside (for example, a user), and may include one or more of a display, a speaker, and the like.
  • the camera 110 is used to collect color images, where the color images collected by the camera are processed by the cyclic edge-preserving smoothing filtering method to obtain an edge-preserving smoothing filtered image.
  • the camera can capture an image desired by the user. (Such as photos, videos, etc.), and then, after the image is processed by the loop edge-preserving smoothing filtering method, an edge-preserving smoothing filtered image is obtained, and the camera may also store the captured image in the memory 104 to For use by other components.
  • the electronic device used to implement the cyclic edge-preserving smoothing filtering method according to the embodiment of the present invention can be implemented as a smart mobile terminal such as a smart phone, a tablet computer, etc., and can also be implemented as any other device with computing capabilities. .
  • an embodiment of a method for cyclic edge-preserving smoothing and filtering is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions And, although the logical sequence is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than here.
  • Fig. 2 is a flowchart of a method for cyclic edge-preserving smoothing filtering according to an embodiment of the present invention.
  • the method is applied to a cyclic edge-preserving smoothing filter (for example, domain transform filter, domain transform filter).
  • the edge-preserving smoothing filter is not specifically limited), as shown in Figure 2, the method includes the following steps:
  • Step S202 Obtain a color image of the target scene and an initial image to be filtered corresponding to the color image.
  • the target scene can be any scene
  • the above-mentioned color image can be an image obtained by shooting the target scene by a camera, a camera, etc., or a pre-stored color image of any scene, or it can be from
  • this embodiment does not specifically limit the foregoing target scene and the manner of obtaining the color image of the target scene.
  • the initial image to be filtered corresponding to the above color image is obtained by performing image segmentation processing on the color image, and is an image that requires edge-preserving smoothing filtering processing.
  • edge-preserving smoothing filtering processing the color image is needed as a guide to assist in correcting the Filter the edges of the original image.
  • step S204 the initial image to be filtered is taken as the first image to be filtered, and the following steps S2041 to S2043 are loop edge-preserving smoothing filtering process performed.
  • This embodiment provides an implementation method for determining the initial value of the loop, including: taking the initial image to be filtered as the first image to be filtered, and the first The image to be filtered is the initial value of the loop.
  • Step S2041 Perform edge-preserving smoothing filtering processing on the current image to be filtered based on the color image to obtain the current smoothing filtered image.
  • the current image to be filtered is subjected to edge-preserving smoothing filtering process based on the color image. If this cycle edge-preserving smoothing filtering process is the first cycle edge-preserving smoothing filtering process, the current image to be filtered is the first image to be filtered obtained in step S204; if this cycle edge-preserving smoothing filtering process is not the first In the loop edge preserving smoothing filtering process, the current image to be filtered is the image output from the last loop edge preserving smoothing filtering process (step S2042 to step S206 will be described in detail).
  • the specific process of performing edge-preserving smoothing filtering processing on the current image to be filtered based on the color image is: performing edge-preserving smoothing filtering processing on each pixel of the current image to be filtered based on the color image.
  • the weight of the filter kernel in the filter is determined by the pixel value of the pixel of the color image corresponding to the A pixel of the current image to be filtered, and then by The filter core determined by the above weight realizes the edge-preserving smoothing filter processing for the A pixel, so that the current smoothed filtered image can be obtained.
  • step S2042 the current image to be filtered is updated based on the current smoothed filtered image and the initial image to be filtered.
  • the current image to be filtered is further updated based on the current smoothing and filtering image and the initial image to be filtered, and the updated image to be filtered is also the image output by this cyclic edge-preserving smoothing filtering process.
  • the process of updating the current image to be filtered will be described in detail below.
  • Step S2043 based on the updated image to be filtered and the current smoothing and filtering image, and the initial image to be filtered, it is determined whether to perform the next cycle edge-preserving smoothing filtering process. If yes, go to step S206; if not, go to step S208.
  • the updated image to be filtered is obtained, based on the updated image to be filtered and the current smoothed filtered image, as well as the initial image to be filtered, decide whether the next cycle edge-preserving smoothing filtering process is needed, that is, whether it can be determined earlier End the process of loop edge preserving smoothing filter processing.
  • the basis for deciding whether to perform the next cycle edge-preserving smoothing filtering process is: after this cycle of edge-preserving smoothing filtering, the obtained updated Whether the image effect of the image to be filtered meets the requirements, it can be specifically based on the first similarity between the updated image to be filtered and the current smoothing filtered image, and the difference between the updated image to be filtered and the initial image to be filtered.
  • the second degree of similarity determines whether the image effect of the updated image to be filtered meets the requirements, and the process will be described in detail below.
  • the above process of determining whether it is necessary to perform the next cycle edge preservation smoothing filter processing is based on the image effect of the output image, which can adapt to the scene of the color image, and can ensure the image effect by reducing the number of loop iterations. Reduce the time-consuming image processing.
  • Step S206 Use the updated image to be filtered as the current image to be filtered, and return to the step of performing edge-preserving smoothing filtering processing on the current image to be filtered based on the color image.
  • Step S208 Use the updated image to be filtered as an edge-preserving smoothing filter processed image obtained by processing the initial image to be filtered by the cyclic edge-preserving smoothing filter.
  • the color image of the target scene and the initial image to be filtered corresponding to the color image are first acquired; then, the initial image to be filtered is taken as the first image to be filtered; the following cyclic edge-preserving smoothing filtering process is performed:
  • the color image performs edge-preserving smoothing filtering processing on the current image to be filtered to obtain the current smoothed filtered image; further, based on the current smoothed filtered image and the initial image to be filtered, the current image to be filtered is updated, and based on the updated image to be filtered Determine whether to perform the next cycle edge-preserving smoothing filtering process with this smoothing and filtering image and the initial image to be filtered.
  • this method will determine whether to perform the next loop edge-preserving smoothing filter based on the updated image to be filtered and the current smooth-filtered image, as well as the initial image to be filtered, each time the loop edge-preserving smoothing filtering process is performed.
  • the method of cyclic edge-preserving smoothing filtering can determine whether the cyclic edge-preserving smoothing filtering process can be ended in advance based on the updated image to be filtered and the current smoothed filtered image, as well as the initial image to be filtered, without reaching the preset
  • the number of cycles that is, the method can adapt to the scene of the color image, reduces the time consumption of image processing, and alleviates the technical problem that the existing loop edge-preserving smoothing filter takes a long time in image processing.
  • a flow chart of a method for cyclic edge-preserving smoothing and filtering, the method of cyclic edge-preserving smoothing and filtering, on the basis of the method shown in Fig. 2, further includes the following steps:
  • step S301 it is judged whether the number of cycles of the cyclic edge preservation smoothing filtering process reaches a preset number of cycles. If the preset number of cycles is reached, step S206 is executed; if the preset number of cycles is not reached, step S2043 is executed.
  • Step S208 Use the updated image to be filtered as an edge-preserving smoothing filter processed image obtained by processing the initial image to be filtered by the cyclic edge-preserving smoothing filter.
  • Step S2043 based on the updated image to be filtered and the current smoothing and filtering image, and the initial image to be filtered, it is determined whether to perform the next cycle edge-preserving smoothing filtering process.
  • the above-mentioned preset number of cycles may be the number of cycles (usually a larger fixed value) that the loop edge-preserving smoothing filter needs to do to achieve the desired effect. From the description of the above content, it can be seen that in the process of each loop edge-preserving smoothing filter processing, if it is determined that the number of loops of the loop-preserving smoothing filter processing reaches the preset number of loops, then the loop is ended; if it is judged that the loop-preserving smoothing filter is smooth If the number of cycles of filtering processing does not reach the preset number of cycles, it is further based on the updated image to be filtered, the current smoothed filtered image, and the initial image to be filtered to determine whether the cycle needs to be terminated early. It can be seen that the method of this embodiment is Under the premise of ensuring the output image effect, the time-consuming image processing can be minimized.
  • step S202 is provided to obtain a color image of the target scene and an implementation manner of the initial image to be filtered corresponding to the color image, including the following steps (1) and (2):
  • image segmentation is usually used to locate objects and boundaries in an image. It is a process of adding labels to each pixel in the image. This process can make pixels with the same label have common visual characteristics.
  • an implementation manner of the foregoing image segmentation processing including:
  • the color image is segmented to obtain the initial image to be filtered corresponding to the color image.
  • image segmentation algorithms can also be used to perform image segmentation processing on the color image, and then obtain the initial image to be filtered corresponding to the color image.
  • the initial image to be filtered can be a grayscale image or a color image.
  • Other image segmentation algorithms can also be: edge detection-based segmentation methods, threshold-based segmentation methods, region-based segmentation methods, etc.
  • the image segmentation algorithm is not specifically limited.
  • step S2042 an implementation manner of updating the current image to be filtered based on the current smoothed filtered image and the initial image to be filtered, including:
  • z i represents the pixel value of the i-th pixel of the updated image to be filtered, Represents the pixel value of the i-th pixel of the smoothed filtered image,
  • t i represents the pixel value of the i-th pixel of the initial image to be filtered, and the value of i is 1, 2, 3,..., M, and M represents The number of pixels, l represents the preset coefficient.
  • the pixel value of each pixel in the updated image to be filtered is the weighted sum of the pixel value of the pixel in the current smoothed filtered image and the initial image to be filtered. It should be noted that the aforementioned pixel value can be any value between 0 and 255.
  • step S2043 is given, based on the updated image to be filtered and the current smoothed filtered image, and the initial image to be filtered to determine whether to perform the next loop edge-preserving smoothing filtering process, refer to Figure 4 includes the following steps:
  • Step S401 Calculate the first similarity between the updated image to be filtered and the current smoothed and filtered image.
  • the image effect of the updated image to be filtered (that is, the image output by a certain cycle edge-preserving smoothing filtering process)
  • it is calculated by calculating the difference between the updated image to be filtered and the current smoothing and filtering image.
  • the first degree of similarity is achieved by calculating the second degree of similarity between the updated image to be filtered and the original image to be filtered.
  • the above-mentioned first similarity measures the degree of edge restoration (that is, edge-preserving smoothing filter processing) with color images. The smaller the value is, the closer the edge of the image outputted by this cycle of edge-preserving smoothing filter processing is to the edge of the color image. , The better the edge repair.
  • the second degree of similarity measures the degree of similarity with the initial image to be filtered. The smaller the value, the closer the image output by this cycle of edge-preserving smoothing filtering processing is to the initial image to be filtered, the better the image effect.
  • the first similarity calculation formula Calculate the first similarity between the updated image to be filtered and the current smoothed filtered image;
  • F 1 represents the first similarity
  • z i represents the pixel value of the i-th pixel of the updated image to be filtered
  • the value of i is 1, 2, 3,..., M, where M represents the number of pixels, and l represents a preset coefficient.
  • the first similarity is obtained by summing up all the pixels between the updated image to be filtered and the current smoothed filtered image of the pixel-to-pixel difference. It should be noted that the aforementioned pixel value can be any value between 0 and 255.
  • Step S402 Calculate the second similarity between the updated image to be filtered and the initial image to be filtered.
  • the second similarity calculation formula Calculate the second similarity between the updated image to be filtered and the initial image to be filtered;
  • F 2 represents the second similarity
  • z i represents the pixel value of the i-th pixel of the updated image to be filtered
  • t i represents The pixel value of the i-th pixel of the initial image to be filtered
  • the value of i is 1, 2, 3,..., M, where M represents the number of pixels.
  • the second similarity is obtained by summing all the pixels between the updated image to be filtered and the original image to be filtered by the square of the pixel-to-pixel difference. It should be noted that the aforementioned pixel value can be any value between 0 and 255.
  • Step S403 Determine whether to perform the next loop edge-preserving smoothing filtering process according to the first similarity and the second similarity.
  • the process of determining whether to perform the next loop edge-preserving smoothing filtering process according to the first similarity and the second similarity includes the following steps:
  • Step S4031 Determine the cost of filtering according to the first similarity and the second similarity. Among them, the filtering cost is used to characterize the filtering effect.
  • the sum of the first similarity degree and the second similarity degree is used as the filtering cost value.
  • the calculation formula of the filtering cost value is: At the same time, it can be seen from the above description that the lower the cost of filtering, the better the filtering effect. In fact, the calculation formula of the filtering cost value can be regarded as a filtering cost function, and the goal is to make the value of the filtering cost function finally smaller than the preset cost threshold (the value can be adjusted as needed).
  • update formula of the image to be filtered is derived by deriving the filtering cost function so that the derivative is equal to zero.
  • step S4032 if the filtering cost value is less than the preset cost threshold, the loop edge-preserving smoothing filtering process is ended.
  • step S4033 if the filtering cost value is not less than the preset cost threshold, perform the next loop edge-preserving smoothing filtering process.
  • the edge-preserving smoothing and filtering image can be used in image processing functions such as background replacement, depth-of-field effect rendering, and portrait blurring.
  • the embodiment of the present invention also provides a device for cyclic edge-preserving smoothing and filtering.
  • the device for cyclic edge-preserving smoothing and filtering is mainly used to implement the method for cyclic edge-preserving smoothing and filtering provided in the above-mentioned embodiments of the present invention.
  • the following is an implementation of the present invention
  • the loop edge preservation smoothing filter device provided in the example is introduced in detail.
  • Fig. 6 is a schematic diagram of a device for cyclic edge preserving smoothing and filtering according to an embodiment of the present invention.
  • the device is applied to a cyclic edge preserving smoothing filter.
  • the device for cyclic edge preserving smoothing and filtering mainly includes: obtaining Unit 10, initialization unit 20, loop edge preservation smoothing filter processing unit 30, wherein:
  • An acquiring unit for acquiring a color image of the target scene and an initial image to be filtered corresponding to the color image
  • the initialization unit is used to use the initial image to be filtered as the first image to be filtered;
  • the loop edge preserving smooth filter processing unit is used to perform the following loop edge preserving smooth filter processing:
  • the color image of the target scene and the initial image to be filtered corresponding to the color image are first acquired; then, the initial image to be filtered is taken as the first image to be filtered; the following cyclic edge-preserving smoothing filtering process is performed:
  • the color image performs edge-preserving smoothing filtering processing on the current image to be filtered to obtain the current smoothed filtered image; further, based on the current smoothed filtered image and the initial image to be filtered, the current image to be filtered is updated, and based on the updated image to be filtered Determine whether to perform the next cycle edge-preserving smoothing and filtering process with this smoothing and filtering image and the initial image to be filtered.
  • this method will determine whether to perform the next cycle edge-preserving smoothing filter based on the updated image to be filtered and the current smooth-filtered image as well as the initial image to be filtered when performing each cycle edge-preserving smoothing filtering process.
  • the method of cyclic edge-preserving smoothing filtering can determine whether the cyclic edge-preserving smoothing filtering process can be terminated in advance based on the updated image to be filtered and the current smoothed filtered image, as well as the initial image to be filtered, without reaching the preset
  • the number of cycles that is, the method can adapt to the scene of the color image, reduces the time consumption of image processing, and alleviates the technical problem that the existing loop edge-preserving smoothing filter takes a long time in image processing.
  • the device is also used to: determine whether the number of cycles of the cyclic edge preservation smoothing filter processing reaches a preset number of cycles; if the number of cycles reaches the preset number of cycles, end the cyclic edge preservation smoothing filter processing, and update the to-be-filtered
  • the image is used as a loop edge-preserving smoothing filter to process the edge-preserving smoothing filtered image obtained from the initial image to be filtered; if the preset number of cycles is not reached, it is based on the updated image to be filtered and the current smoothing filtered image, as well as the initial image to be filtered Determine whether to perform the next loop edge-preserving smoothing filter processing.
  • the above-mentioned acquisition unit is further configured to: acquire a color image of the target scene; perform image segmentation processing on the color image to obtain an initial image to be filtered corresponding to the color image.
  • the aforementioned loop edge-preserving smoothing filtering processing unit is further configured to: according to the update formula of the image to be filtered Update the current image to be filtered to obtain the updated image to be filtered; z i represents the pixel value of the i-th pixel of the updated image to be filtered, Represents the pixel value of the i-th pixel of the smoothed filtered image, t i represents the pixel value of the i-th pixel of the initial image to be filtered, and the value of i is 1, 2, 3,..., M, and M represents The number of pixels, l represents the preset coefficient.
  • the aforementioned cyclic edge-preserving smoothing filtering processing unit is further used to: calculate the first similarity between the updated image to be filtered and the current smoothed and filtered image; calculate the difference between the updated image to be filtered and the initial image to be filtered The second degree of similarity between the two; according to the first degree of similarity and the second degree of similarity, determine whether to perform the next loop edge-preserving smoothing filtering process.
  • the above-mentioned circular edge-preserving smoothing filter processing unit is further configured to: calculate the formula according to the first similarity degree Calculate the first similarity between the updated image to be filtered and the current smoothed filtered image;
  • F 1 represents the first similarity
  • z i represents the pixel value of the i-th pixel of the updated image to be filtered
  • the value of i is 1, 2, 3,..., M, where M represents the number of pixels, and l represents a preset coefficient.
  • the above-mentioned circular edge-preserving smoothing filtering processing unit is further configured to: calculate the formula according to the second similarity degree Calculate the second similarity between the updated image to be filtered and the initial image to be filtered;
  • F 2 represents the second similarity
  • z i represents the pixel value of the i-th pixel of the updated image to be filtered
  • t i represents The pixel value of the i-th pixel of the initial image to be filtered
  • the value of i is 1, 2, 3,..., M, where M represents the number of pixels.
  • the aforementioned loop edge-preserving smoothing filter processing unit is further configured to: determine the filtering cost value according to the first similarity and the second similarity; the filtering cost value is used to characterize the filtering effect; if the filtering cost value is less than the preset cost threshold, The loop edge-preserving smoothing filtering process is ended; if the cost of filtering is not less than the preset cost threshold, the next loop-preserving smoothing filtering process is performed.
  • the aforementioned loop edge-preserving smoothing filtering processing unit is further configured to: use the sum of the first similarity degree and the second similarity degree as the filtering cost value.
  • a computer storage medium on which a computer program is stored, and the computer executes the steps of the method described in any one of the above method embodiments when the computer program runs the computer program.
  • the device embodiments described above are merely illustrative, where the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments. Those of ordinary skill in the art can understand and implement without creative work.
  • the various component embodiments of the present invention may be implemented by hardware, or by software modules running on one or more processors, or by a combination of them.
  • a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the computing processing device according to the embodiments of the present invention.
  • DSP digital signal processor
  • the present invention can also be implemented as a device or device program (for example, a computer program and a computer program product) for executing part or all of the methods described herein.
  • Such a program for realizing the present invention may be stored on a computer-readable medium, or may have the form of one or more signals.
  • Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.
  • FIG. 7 shows a computing processing device that can implement the method according to the present invention.
  • the computing processing device traditionally includes a processor 1010 and a computer program product in the form of a memory 1020 or a computer readable medium.
  • the memory 1020 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read Only Memory), EPROM, hard disk, or ROM.
  • the memory 1020 has a storage space 1030 for executing program codes 1031 of any method steps in the above methods.
  • the storage space 1030 for program codes may include various program codes 1031 respectively used to implement various steps in the above method. These program codes can be read from or written into one or more computer program products.
  • These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards, or floppy disks. Such computer program products are usually portable or fixed storage units as described with reference to FIG. 8.
  • the storage unit may have storage segments, storage spaces, etc. arranged similarly to the memory 1020 in the computing processing device of FIG. 7.
  • the program code can be compressed in an appropriate form, for example.
  • the storage unit includes computer-readable codes 1031', that is, codes that can be read by, for example, a processor such as 1010. These codes, when run by a computing processing device, cause the computing processing device to execute the method described above. The various steps.
  • the terms “installed”, “connected”, and “connected” should be understood in a broad sense, for example, they may be fixed connections or detachable connections. , Or integrally connected; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication between two components.
  • installed e.g., they may be fixed connections or detachable connections. , Or integrally connected; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication between two components.
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation.
  • multiple units or components may be combined or It can be integrated into 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 through some communication interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, 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 invention 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 function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a nonvolatile computer readable storage medium executable by a processor.
  • the technical solution of the present invention essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present invention.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes. .
  • any reference signs placed between parentheses should not be constructed as a limitation to the claims.
  • the word “comprising” does not exclude the presence of elements or steps not listed in the claims.
  • the word “a” or “an” preceding an element does not exclude the presence of multiple such elements.
  • the invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims that list several devices, several of these devices may be embodied in the same hardware item.
  • the use of the words first, second, and third, etc. do not indicate any order. These words can be interpreted as names.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Facsimile Image Signal Circuits (AREA)

Abstract

L'invention concerne un procédé et un appareil de filtration lisse préservant les bords cycliques, et un dispositif électronique. Le procédé consiste : à acquérir une image couleur d'un scénario cible et une image initiale à filtrer correspondant à l'image couleur (S202) ; et à amener l'image initiale à filtrer en tant que première image à filtrer (S204), et à réaliser un traitement de filtration lisse préservant les bords cycliques comme suit : sur la base de l'image couleur, à réaliser un traitement de filtrage lisse préservant les bords sur l'image courante à filtrer afin d'obtenir la présente image filtrée lisse (S2041) ; sur la base de la présente image filtrée lisse et de l'image initiale à filtrer, à mettre à jour l'image courante à filtrer (S2042) ; et sur la base de l'image mise à jour à filtrer, de la présente image filtrée lisse, et de l'image initiale à filtrer, à déterminer s'il faut effectuer le traitement de filtration lisse préservant les bords cycliques suivants (S2043). Selon le procédé, il peut être déterminé, en fonction de l'image mise à jour à filtrer, de la présente image filtrée lisse et de l'image initiale à filtrer, si le traitement de filtration lisse préservant le bord cyclique peut être terminé à l'avance, sans avoir besoin d'atteindre un nombre prédéfini de cycles, c'est-à-dire, le procédé peut être auto-adapté au scénario de l'image couleur, ce qui permet de réduire le temps consommé pour le traitement d'image.
PCT/CN2020/097641 2019-08-29 2020-06-23 Procédé et appareil de filtration lisse préservant les bords cycliques, et dispositif électronique WO2021036442A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910811227.7 2019-08-29
CN201910811227.7A CN110517201B (zh) 2019-08-29 2019-08-29 循环保边平滑滤波的方法、装置和电子设备

Publications (1)

Publication Number Publication Date
WO2021036442A1 true WO2021036442A1 (fr) 2021-03-04

Family

ID=68629421

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/097641 WO2021036442A1 (fr) 2019-08-29 2020-06-23 Procédé et appareil de filtration lisse préservant les bords cycliques, et dispositif électronique

Country Status (2)

Country Link
CN (1) CN110517201B (fr)
WO (1) WO2021036442A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113298761A (zh) * 2021-05-07 2021-08-24 奥比中光科技集团股份有限公司 一种图像滤波方法、装置、终端和计算机可读存储介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110517201B (zh) * 2019-08-29 2021-11-16 北京迈格威科技有限公司 循环保边平滑滤波的方法、装置和电子设备
CN111314579B (zh) * 2020-02-21 2021-10-15 苏州浪潮智能科技有限公司 一种图像垂直滤波处理的方法及芯片

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103391446A (zh) * 2013-06-24 2013-11-13 南京大学 一种基于自然场景统计的深度图优化方法
CN103400359A (zh) * 2013-08-07 2013-11-20 中国科学院长春光学精密机械与物理研究所 基于非局部域变换的实时彩色图像滤波方法
US20170109896A1 (en) * 2015-10-15 2017-04-20 Samsung Electronics, Co., Ltd. Method of measuring thickness, method of processing image and electronic system performing the same
CN106920222A (zh) * 2017-03-13 2017-07-04 苏州大学 一种图像平滑方法及装置
CN110517201A (zh) * 2019-08-29 2019-11-29 北京迈格威科技有限公司 循环保边平滑滤波的方法、装置和电子设备

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9005867B2 (en) * 2013-08-07 2015-04-14 Xerox Corporation Porous toner and process for making the same
CN104715457A (zh) * 2015-03-19 2015-06-17 天津大学 基于边缘保护的快速加权各向异性扩散滤波方法
CN105631825B (zh) * 2015-12-28 2018-11-16 西安电子科技大学 基于滚动导向的图像去雾方法
CN109389565A (zh) * 2018-10-19 2019-02-26 山东大学 一种保持边缘和平滑度可调节的图像平滑方法
CN109711466B (zh) * 2018-12-26 2023-04-14 陕西师范大学 一种基于边缘保留滤波的cnn高光谱图像分类方法
CN109919855A (zh) * 2019-01-17 2019-06-21 深圳禾苗通信科技有限公司 一种基于多帧影像的快速保边去噪方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103391446A (zh) * 2013-06-24 2013-11-13 南京大学 一种基于自然场景统计的深度图优化方法
CN103400359A (zh) * 2013-08-07 2013-11-20 中国科学院长春光学精密机械与物理研究所 基于非局部域变换的实时彩色图像滤波方法
US20170109896A1 (en) * 2015-10-15 2017-04-20 Samsung Electronics, Co., Ltd. Method of measuring thickness, method of processing image and electronic system performing the same
CN106920222A (zh) * 2017-03-13 2017-07-04 苏州大学 一种图像平滑方法及装置
CN110517201A (zh) * 2019-08-29 2019-11-29 北京迈格威科技有限公司 循环保边平滑滤波的方法、装置和电子设备

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113298761A (zh) * 2021-05-07 2021-08-24 奥比中光科技集团股份有限公司 一种图像滤波方法、装置、终端和计算机可读存储介质

Also Published As

Publication number Publication date
CN110517201B (zh) 2021-11-16
CN110517201A (zh) 2019-11-29

Similar Documents

Publication Publication Date Title
WO2021036442A1 (fr) Procédé et appareil de filtration lisse préservant les bords cycliques, et dispositif électronique
US11373275B2 (en) Method for generating high-resolution picture, computer device, and storage medium
WO2019134504A1 (fr) Procédé et dispositif de floutage d'arrière-plan d'image, support de stockage et appareil électronique
CN109308463B (zh) 一种视频目标识别方法、装置及设备
WO2021068330A1 (fr) Procédé et dispositif de segmentation et de classification d'image intelligente et support d'informations lisible par ordinateur
JP2015225665A (ja) 画像ノイズ除去方法及び画像ノイズ除去装置
CN108337551B (zh) 一种屏幕录制方法、存储介质及终端设备
CN108230346B (zh) 用于分割图像语义特征的方法和装置、电子设备
WO2016101757A1 (fr) Procédé et dispositif de traitement d'image basés sur un dispositif mobile
CN109376256B (zh) 图像搜索方法及装置
CN108961267B (zh) 图片处理方法、图片处理装置及终端设备
CN110335216B (zh) 图像处理方法、图像处理装置、终端设备及可读存储介质
CN111666960A (zh) 图像识别方法、装置、电子设备及可读存储介质
CN111008935B (zh) 一种人脸图像增强方法、装置、系统及存储介质
KR102580519B1 (ko) 영상처리장치 및 기록매체
CN110889824A (zh) 一种样本生成方法、装置、电子设备及计算机可读存储介质
WO2023098045A1 (fr) Procédé et appareil d'alignement d'image, et dispositif informatique et support de stockage
WO2021114843A1 (fr) Procédé et appareil de traitement d'image d'empreinte digitale, dispositif électronique et support lisible par ordinateur
CN112991180A (zh) 图像拼接方法、装置、设备以及存储介质
CN108229583B (zh) 一种基于主方向差分特征的快速模板匹配的方法及装置
WO2018058476A1 (fr) Procédé et dispositif de correction d'image
CN110503704B (zh) 三分图的构造方法、装置和电子设备
CN108932703B (zh) 图片处理方法、图片处理装置及终端设备
CN108734718B (zh) 用于图像分割的处理方法、装置、存储介质及设备
CN108875538B (zh) 属性检测方法、装置和系统及存储介质

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: 20856245

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: 20856245

Country of ref document: EP

Kind code of ref document: A1