WO2022002002A1 - Procédé de traitement d'image, appareil de traitement d'image, dispositif électronique et support de stockage - Google Patents

Procédé de traitement d'image, appareil de traitement d'image, dispositif électronique et support de stockage Download PDF

Info

Publication number
WO2022002002A1
WO2022002002A1 PCT/CN2021/102912 CN2021102912W WO2022002002A1 WO 2022002002 A1 WO2022002002 A1 WO 2022002002A1 CN 2021102912 W CN2021102912 W CN 2021102912W WO 2022002002 A1 WO2022002002 A1 WO 2022002002A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
binarized
pixel
pixels
binarization
Prior art date
Application number
PCT/CN2021/102912
Other languages
English (en)
Chinese (zh)
Inventor
徐青松
李青
Original Assignee
杭州睿琪软件有限公司
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 杭州睿琪软件有限公司 filed Critical 杭州睿琪软件有限公司
Publication of WO2022002002A1 publication Critical patent/WO2022002002A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/73—Deblurring; Sharpening
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • 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
    • G06T2207/10004—Still image; Photographic 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
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging

Definitions

  • the present invention relates to the technical field of digital image processing, and in particular, to an image processing method, an image processing apparatus, an electronic device, and a non-transitory computer-readable storage medium.
  • image binarization processing is to set the gray value of pixels on the image to 0 or 255, and image binarization processing can greatly reduce the amount of data in the image. , so that the outline of the target of interest can be highlighted, and in addition, it is convenient to process and analyze the image, for example, to extract the information in the image.
  • At least one embodiment of the present disclosure provides an image processing method, including: acquiring an original image, wherein the original image includes at least one object; and processing the original image by using a first binarization model to obtain the original image
  • the first binarized image of the image is processed to obtain a pixel circumscribed contour image, wherein the pixel circumscribed contour image includes a plurality of circumscribed contour pixels, and the plurality of circumscribed contour pixels
  • the pixels in the enclosed area are pixels corresponding to at least part of the objects in the at least one object
  • the original image is processed by a second binarization model to obtain a second binarized image; according to the plurality of objects
  • the position of the circumscribed contour pixel in the pixel circumscribed contour image, and the second binarized image and the first binarized image are synthesized to obtain a synthesized image, wherein the synthesized image is a binary image image.
  • the original image is processed by the first binarization model to obtain the first binarized image of the original image, including: The original image is compressed to obtain an input image, wherein the size of the input image is smaller than the size of the original image; and the input image is processed by the first binarization model to obtain the The first binarized image of the original image.
  • processing the input image by using the first binarization model to obtain the first binarized image of the original image includes: The first binarization model processes the input image to obtain a binarized predicted image of the input image; and restores the size of the binarized predicted image to obtain the first binarized image
  • the size of the first binarized image is the same as the size of the original image.
  • processing the first binarized image to obtain a pixel circumscribed contour image includes: performing blurring processing on the first binarized image , obtain a blurred image; perform XOR processing on the blurred image and the first binarized image to obtain the pixel circumscribed contour image.
  • the second binarized image and the first binary image are analyzed.
  • Synthesizing the binarized images to obtain the synthesized image includes: acquiring the positions of the plurality of circumscribed contour pixels in the pixel circumscribed contour image; extracting the plurality of circumscribed contour pixels in the second binarized image multiple target second binarized pixels at positions corresponding to the positions of the contour pixels; according to the pixel correspondence between the second binarized image and the first binarized image, the second binarized image is The plurality of target second binarized pixels in the image are respectively combined into the same position in the first binarized image to obtain the combined image.
  • all the second binarized pixels in the second binarized image are arranged in n rows and m columns, and the pixels in the first binarized image are arranged in n rows and m columns.
  • All the first binarized pixels are arranged in n rows and m columns, and all the synthesized pixels in the composite image are arranged in n rows and m columns, according to the pixels of the second binarized image and the first binarized image.
  • synthesizing the multiple target second binarized pixels in the second binarized image to the same position in the first binarized image respectively includes: determining the multiple target second binarized pixels.
  • the q th target second binarized pixel in the two binarized pixels wherein the q th target second binarized pixel is located in the ith row and the jth column of the second binarized image; determine The q-th target first binarized pixel located in the i-th row and j-th column in the first binarized image; replace the grayscale value of the q-th target first binarized pixel with the grayscale value of the second binarized pixel of the q th target, so as to obtain the q th target composite pixel located in the i th row and the j th column in the composite image, wherein the q th target composite pixel
  • the grayscale value of the pixel is the grayscale value of the qth target second binarized pixel, n, m, q, i, and j are all positive integers, and i is less than or equal to n, j is less than or equal to m, and q is less than or equal to is
  • processing the original image by using a second binarization model to obtain a second binarized image includes: performing grayscale on the original image process the grayscale image to obtain a grayscale image; process the grayscale image according to the first threshold to obtain an intermediate binarized image; use the intermediate binarized image as a guide map to perform guided filtering on the grayscale image process to obtain a filtered image; determine high-value pixels in the filtered image according to a second threshold, wherein the gray value of the high-value pixels is greater than the second threshold; The gray value of the high-value pixels is expanded to obtain an expanded image; the expanded image is sharpened to obtain a clear image; and the contrast of the clear image is adjusted to obtain the second binarization image.
  • the first binarization model is a model based on a neural network.
  • At least one embodiment of the present disclosure provides an image processing apparatus, including: an acquisition module for acquiring an original image, wherein the original image includes at least one object; and a first binarization module for obtaining an original image through a first binarization
  • the model processes the original image to obtain a first binarized image of the original image;
  • a processing module is used to process the first binarized image to obtain a pixel circumscribed contour image, wherein the The pixel circumscribed contour image includes a plurality of circumscribed contour pixels, and the pixels in the area surrounded by the plurality of circumscribed contour pixels are pixels corresponding to at least part of the objects in the at least one object;
  • the second binarization module is configured to pass
  • the second binarization model processes the original image to obtain a second binarized image;
  • the synthesis module is configured to, according to the positions of the plurality of circumscribed contour pixels in the pixel circumscribed contour image, analyze the The second binarized image and the first binarized image are combined to obtain the combined image, wherein
  • the first binarization module performs processing on the original image through a first binarization model, so as to obtain the first two images of the original image.
  • the operation of binarizing the image includes performing the following operations: compressing the original image to obtain an input image, wherein the size of the input image is smaller than the size of the original image;
  • the model processes the input image to obtain a first binarized image of the original image.
  • the first binarization module performs processing on the input image by using the first binarization model, so as to obtain a second image of the original image.
  • the operation of a binarized image includes performing the following operations: processing the input image by using the first binarization model to obtain a binarized predicted image of the input image; The predicted image is resized to obtain the first binarized image, wherein the size of the first binarized image is the same as the size of the original image.
  • the processing module when the processing module performs the operation of processing the first binarized image to obtain a pixel circumscribed contour image, the operation includes performing the following operations: performing a blurring process on the first binarized image to obtain a blurred image; and performing XOR processing on the blurred image and the first binarized image to obtain the pixel circumscribed contour image.
  • the synthesizing module performs, according to the positions of the plurality of circumscribed contour pixels in the pixel circumscribed contour image, the second binarized image and the
  • the operation includes performing the following operations: obtaining the positions of the plurality of circumscribed contour pixels in the pixel circumscribed contour image; multiple target second binarized pixels at positions corresponding to the positions of the plurality of circumscribed contour pixels in the second binarized image; according to the second binarized image and the first binarized
  • the pixel correspondence of the image, the multiple target second binarized pixels in the second binarized image are respectively synthesized into the same position in the first binarized image to obtain the original image composite image.
  • all the second binarized pixels in the second binarized image are arranged in n rows and m columns, and the pixels in the first binarized image are arranged in n rows and m columns.
  • All the first binarized pixels are arranged in n rows and m columns, and all the synthesized pixels in the composite image are arranged in n rows and m columns, and the composite module executes the process according to the second binarized image and the first binarized image.
  • the pixel correspondence relationship of the binarized image, and the operation of synthesizing the plurality of target second binarized pixels in the second binarized image to the same position in the first binarized image respectively includes: performing the following operations: determining the qth target second binarization pixel in the plurality of target second binarization pixels, wherein the qth target second binarization pixel is located in the second binarization pixel The i-th row and the j-th column of the binarized image are determined; the q-th target first binarized pixel located in the i-th row and the j-th column in the first binarized image is determined; the q-th target first binarized pixel is The gray-level value of a binarized pixel is replaced with the gray-level value of the q-th target second binarized pixel, so as to obtain the q-th target composite in the i-th row and j-th column in the composite image pixel, wherein the grayscale value of the qth target
  • the second binarization module includes: a grayscale module, configured to perform grayscale processing on the original image to obtain a grayscale image; an intermediate The binarization module is used to process the grayscale image according to the first threshold to obtain an intermediate binarized image; the filtering module is used to take the intermediate binarized image as a guide map, and to process the grayscale image.
  • a grayscale module configured to perform grayscale processing on the original image to obtain a grayscale image
  • an intermediate The binarization module is used to process the grayscale image according to the first threshold to obtain an intermediate binarized image
  • the filtering module is used to take the intermediate binarized image as a guide map, and to process the grayscale image.
  • the image is subjected to guided filtering processing to obtain a filtered image; a determination module is configured to determine high-value pixels in the filtered image according to a second threshold, wherein the gray value of the high-value pixels is greater than the second threshold
  • the expansion module is used to expand the gray value of the high-value pixel points according to the preset expansion coefficient to obtain an expanded image; the clearing module is used to clear the expanded image to obtain a clear image. , and adjust the contrast of the clear image to obtain the second binarized image.
  • At least one embodiment of the present disclosure provides an electronic device, comprising: a memory for non-transitory storage of computer-readable instructions; a processor for executing the computer-readable instructions, the computer-readable instructions being executed by the The image processing method according to any embodiment of the present disclosure is implemented when the processor runs.
  • At least one embodiment of the present disclosure provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer-readable instructions that, when executed by a processor, implement a The image processing method described in any embodiment of the present disclosure.
  • FIG. 1 is a schematic flowchart of an image processing method provided by at least one embodiment of the present disclosure
  • FIG. 2 is a schematic diagram of an original image provided by at least one embodiment of the present disclosure
  • FIG. 3 shows an input image provided by an embodiment of the present disclosure
  • FIG. 4 is a schematic diagram of a first binarized image provided by an embodiment of the present disclosure.
  • FIG. 5 is a schematic diagram of a blurred image provided by an embodiment of the present disclosure.
  • FIG. 6 is a pixel circumscribed contour image provided by an embodiment of the present disclosure.
  • FIG. 7 is a schematic diagram of a second binarized image according to an embodiment of the present disclosure.
  • FIG. 8 is a schematic diagram of a composite image provided by an embodiment of the present disclosure.
  • FIG. 9 is a schematic block diagram of an image processing apparatus according to at least one embodiment of the present disclosure.
  • FIG. 10 is a schematic block diagram of an electronic device according to at least one embodiment of the present disclosure.
  • FIG. 11 is a schematic diagram of a non-transitory computer-readable storage medium provided by at least one embodiment of the present disclosure.
  • FIG. 12 is a schematic diagram of a hardware environment provided by at least one embodiment of the present disclosure.
  • At least one embodiment of the present disclosure provides an image processing method, an image processing apparatus, an electronic device, and a non-transitory computer-readable storage medium.
  • the image processing method includes: acquiring an original image, wherein the original image includes at least one object; processing the original image through a first binarization model to obtain a first binarized image of the original image; Perform processing to obtain a pixel circumscribed contour image, wherein the pixel circumscribed contour image includes a plurality of circumscribed contour pixels, and the pixels in the area surrounded by the plurality of circumscribed contour pixels are pixels corresponding to at least part of the at least one object;
  • the binarization model processes the original image to obtain a second binarized image; the second binarized image and the first binarized image are synthesized according to the positions of multiple circumscribed contour pixels in the pixel circumscribed contour image , to get a composite image of the original image.
  • the composite image is a binarized image.
  • the image processing method can convert an original image (for example, a color image or an unclear image) into a binarized image with obvious and clear black and white contrast, which can effectively improve the quality of the binarized image and improve the recognition degree of the image content.
  • an original image for example, a color image or an unclear image
  • the converted image has less noise and obvious black and white contrast, it can effectively improve the printing effect.
  • the image processing method provided by the embodiment of the present disclosure can be applied to the image processing apparatus provided by the embodiment of the present disclosure, and the image processing apparatus can be configured on an electronic device.
  • the electronic device may be a personal computer, a mobile terminal, etc.
  • the mobile terminal may be a hardware device with various operating systems, such as a mobile phone, a tablet computer, and the like.
  • FIG. 1 is a schematic flowchart of an image processing method provided by at least one embodiment of the present disclosure
  • FIG. 2 is a schematic diagram of an original image provided by at least one embodiment of the present disclosure.
  • the image processing method provided by the embodiment of the present disclosure includes steps S10 to S14.
  • the image processing method includes step S10 : acquiring an original image.
  • the original image includes at least one object
  • the object may be a character
  • the character may include Chinese (for example, Chinese characters or pinyin), English, Japanese, French, Korean, Latin, numbers, etc.
  • the object may also include various symbols ( For example, greater than sign, less than sign, percent sign, etc.) and various graphics, etc.
  • At least one of the objects may include printed or machine-typed characters, as well as handwritten characters.
  • objects in the original image may include words and letters in print (eg, English, Japanese, French, Korean, German, Latin, etc. in different national languages and scripts) ), printed numbers (eg, dates, weights, dimensions, etc.), printed symbols and images, etc., handwritten words and letters, handwritten numbers, handwritten symbols and graphics, etc.
  • the original image may be various types of images, for example, may be an image of a shopping list, an image of a dining receipt, an image of a test paper, an image of a contract, and the like. As shown in FIG. 2, the original image may be an image of a letter.
  • the shape of the original image can be rectangle etc.
  • the shape and size of the original image can be set by the user according to the actual situation.
  • the original image may be an image captured by an image acquisition device (eg, a digital camera or a mobile phone, etc.), and the original image may be a grayscale image or a color image.
  • the original image refers to a form in which the object to be processed (eg, test paper, cooperation, shopping receipt, etc.) is presented in a visual manner, such as a picture of the object to be processed.
  • the original image can also be obtained by scanning or the like.
  • the original image may be an image directly collected by an image collection device, or an image obtained after preprocessing the collected image.
  • the image processing method provided by at least one embodiment of the present disclosure may further include an operation of preprocessing the original image.
  • the preprocessing may include, for example, processing, such as cropping, gamma (Gamma) correction, or noise reduction filtering, on the image directly collected by the image collection device. Preprocessing can eliminate irrelevant information or noise information in the original image, so as to facilitate better image processing of the original image.
  • step S11 the original image is processed through a first binarization model to obtain a first binarized image of the original image.
  • the first binarization model is a neural network-based model.
  • the first binarization model may be implemented using machine learning techniques and run, for example, on a general purpose computing device or a special purpose computing device.
  • the first binarization model is a neural network model obtained by pre-training.
  • the first binarization model can be implemented by using a U-net neural network, a neural network similar to the U-net neural network, a Mask-rcnn neural network, or other neural networks.
  • the first binarization model is used to binarize the original image to obtain the first binarized image.
  • the binarization process is to set the gray value of the pixel on the original image to the first gray value (for example, 0) or the second gray value (for example, 255), that is, to make the whole original image appear obvious Process of black and white effect.
  • the first binarization model to be trained can be trained through a large number of original training images and the binarized images of the original training images, and then the first binarization model (for example, a neural network model such as a U-net neural network) is established. ).
  • the first binarization model for example, a neural network model such as a U-net neural network
  • an existing binarization processing method may also be used to perform binarization processing on the original image to obtain the first binarized image.
  • the binarization processing method may include a threshold value method, and the threshold value method includes: setting a binarization threshold value, and comparing the gray level value of each pixel in the original image with the binarization threshold value. If the grayscale value is greater than or equal to the binarization threshold, the grayscale value of the pixel is set to 255 grayscale. If the grayscale value of a pixel in the original image is less than the binarization threshold, the grayscale value of the pixel is set to Set it to 0 grayscale, so that the original image can be binarized.
  • the selection method of the binarization threshold includes the bimodal method, the P-parameter method, the big law (OTSU method), the maximum entropy method, and the iterative method.
  • step S11 may include: compressing the original image to obtain an input image; and processing the input image through a first binarization model to obtain a first binarized image of the original image.
  • FIG. 3 shows an input image provided by an embodiment of the present disclosure.
  • the input image in FIG. 3 may be an image obtained by compressing the original image in FIG. 2 .
  • the dimensions of the input image are smaller than the dimensions of the original image. It should be noted that when the first binarization model is used for the binarization prediction processing, if the original image is too large, the processing speed will be relatively slow. Therefore, in order to improve the processing speed, the original image can be compressed to obtain The compressed input image is then subjected to binarization prediction processing on the input image through the first binarization model.
  • the size of the input image may be 1500*1500. It should be noted that the size of the input image is not limited to this, and can be set according to the actual situation.
  • the user can preset the compression ratio, etc.
  • the compression ratio is related to factors such as processing speed and image quality, and the user can comprehensively consider the processing speed and image quality. factor to set the compression ratio. If the processing speed needs to be made faster, the compression ratio can be larger, so that the size of the input image is smaller. However, at this time, the quality of the final first binarized image may be poor; if the final second binarized image needs to be obtained If the quality of the binarized image is better, the compression ratio can be smaller, so that the size of the input image is larger.
  • the processing speed is slower, that is, the first binarization model performs binarization processing on the input image.
  • process is longer.
  • the aspect ratio of the input image and the aspect ratio of the original image may be the same. In other examples, the aspect ratio of the input image and the aspect ratio of the original image may be different. In this case, relative to the original image , the input image is deformed. After the input image is processed, a binarized predicted image is obtained. The size of the binarized predicted image needs to be restored to the same size as the original image.
  • the resolution of the photo is 12 million pixels, that is, 4000*3000. At this time, the size of the original image can be 4000*3000, and the size of the input image obtained after compressing the original image can also be 1500*1500. This , the input image is deformed.
  • FIG. 4 is a schematic diagram of a first binarized image provided by an embodiment of the present disclosure.
  • processing the input image through the first binarization model to obtain the first binarization image of the original image includes: processing the input image through the first binarization model to obtain the first binarization model of the input image. Binarize the predicted image; restore the size of the binarized predicted image to obtain a first binarized image.
  • the first binarized image in FIG. 4 is the binarized image of the original image shown in FIG. 2 .
  • the size of the first binarized image is the same as the size of the original image.
  • the black pixels represent the pixels corresponding to the object
  • the white pixels represent the pixels corresponding to the background.
  • the size of the binarized predicted image is the same as the size of the input image.
  • the size of the binarized predicted image is restored, that is, the size of the binarized predicted image is enlarged according to the enlargement ratio, so that the size of the first binarized image is equal to the size of the first binarized image.
  • the size of the original image is the same, so that it is convenient to perform pixel synthesis at the corresponding position later.
  • the enlargement ratio may correspond to the compression ratio. For example, if the compression ratio is 1/K, the enlargement ratio may be K.
  • an interpolation method may be used to restore the size of the binarized predicted image to obtain the first binarized image.
  • FIG. 5 is a schematic diagram of a blurred image according to an embodiment of the disclosure
  • FIG. 6 is a pixel circumscribed contour image according to an embodiment of the disclosure.
  • step S12 the first binarized image is processed to obtain a pixel circumscribed contour image.
  • the pixel circumscribed contour image shown in FIG. 6 is a pixel circumscribed contour image obtained by processing the first binarized image shown in FIG. 4 .
  • the pixel circumscribed contour image includes a plurality of circumscribed contour pixels, and the white pixels in FIG. 6 represent circumscribed contour pixels.
  • the pixels in the area surrounded by the plurality of circumscribed contour pixels are pixels corresponding to at least part of the at least one object, and the black pixels inside the white pixels in FIG. 6 represent the pixels of the object.
  • step S12 includes: performing a blurring process on the first binarized image to obtain a blurred image; and performing XOR processing on the blurred image and the first binarized image to obtain a pixel circumscribed contour image.
  • the blurred image shown in FIG. 5 may be a blurred image obtained by blurring the first binarized image shown in FIG. 4 .
  • the mask area (Mask area) of the object in the first binarized image becomes larger.
  • Gaussian filtering may be used to blur the first binarized image. It should be noted that, in the present disclosure, the method of fuzzification processing is not limited to Gaussian filtering, and may also be other suitable methods, such as median filtering, mean filtering, and the like.
  • the white pixels in Fig. 6 represent different pixels between the blurred image and the first binarized image, that is, for any white pixel in Fig. 6, the pixel at the position corresponding to the white pixel in the blurred image
  • the grayscale value is different from the grayscale value of the pixel at the position corresponding to the white pixel in the first binarized image.
  • the black pixels in Fig. 6 represent the same pixels between the blurred image and the first binarized image, that is, for any black pixel in Fig. 6, the pixel at the position corresponding to the black pixel in the blurred image
  • the grayscale value is the same as the grayscale value of the pixel at the position corresponding to the black pixel in the first binarized image.
  • FIG. 7 is a schematic diagram of a second binarized image according to an embodiment of the present disclosure.
  • step S13 the original image is processed through a second binarization model to obtain a second binarized image.
  • the second binarized image shown in FIG. 7 is an image obtained by processing the original image shown in FIG. 2 through the second binarization model.
  • step S13 the processing performed by the second binarization model (for example, lessink processing) is performed according to the original image.
  • the processing performed by the second binarization model can be used to remove part of the grayscale in the original image. pixels, while enhancing the detail information of objects (eg, characters), that is, more detailed pixel features can be preserved.
  • the processing performed by the second binarization model can also remove image noise interference in the original image, making the details of the object more prominent.
  • the binarized prediction image obtained by performing the binarization prediction process on the input image shown in Figure 3 does not perform well in the detail part and has jaggedness. Therefore, it is necessary to supplement the detail pixels of the image, so that the final composite image has better effect.
  • step S13 may include: performing grayscale processing on the original image to obtain a grayscale image; processing the grayscale image according to the first threshold to obtain an intermediate binarized image; As a guide map, conduct guide filtering processing on the grayscale image to obtain a filtered image; determine high-value pixels in the filtered image according to a second threshold, wherein the grayscale value of the high-value pixels is greater than the second threshold; according to a preset The expansion coefficient is used to expand the gray value of the high-value pixels to obtain an expanded image; to clear the expanded image to obtain a clear image; and to adjust the contrast of the clear image to obtain a second binarized image.
  • grayscale processing methods include component method, maximum value method, average value method, weighted average method, and the like.
  • a threshold method can be used to binarize a grayscale image to obtain an intermediate binarized image.
  • the commonly used binarization threshold selection methods include the bimodal method, the P parameter method, the big law (OTSU method), the maximum entropy method, the iterative method, etc.
  • the selection method of the first threshold can be any of the above methods. kind.
  • the first threshold can be set according to the actual situation, which is not specifically limited here.
  • step S13 in the guided filtering process, the intermediate binarized image is used as the guided image, the grayscale image is used as the input image in the guided filtering process, and the filtered image is the output image in the guided filtering process.
  • the valued image performs guided filtering on the grayscale image, which can output a filtered image that is roughly similar to the grayscale image and similar to the intermediate binarized image at the edge texture. After the guided filtering process, the noise in the image is significantly reduced.
  • the second threshold is the sum of the average gray value of the filtered image and the standard deviation of the gray value, that is, the second threshold is equal to the average value of the gray value of each pixel in the filtered image plus the gray value of each pixel in the filtered image.
  • the standard deviation of the degree value is the standard deviation of the degree value.
  • the preset expansion factor is 1.2-1.5, eg, 1.3.
  • the gray value of each high-value pixel is multiplied by a preset expansion coefficient to perform expansion processing on the gray value of the high-value pixel, thereby obtaining an expanded image with more obvious black and white contrast.
  • performing a sharpening process on the expanded image to obtain a clear image includes: using Gaussian filtering to perform blurring processing on the expanded image to obtain a blurred image corresponding to the expanded image; according to a preset mixing coefficient, The blurred image corresponding to the expanded image and the expanded image are mixed proportionally to obtain a clear image.
  • f 1 (i, j) is the gray value of the pixel point at (i, j) of the extended image
  • f 2 (i, j) is the pixel point of the blurred image corresponding to the extended image at (i, j)
  • f 3 (i, j) is the gray value of the pixel point at (i, j) of the clear image
  • k 1 is the preset mixing coefficient of the extended image
  • k 2 is the blurred image corresponding to the extended image
  • f 3 (i,j) k 1 f 1 (i,j)+k 2 f 2 (i,j).
  • the preset mixing coefficient of the extended image may be 1.5, and the preset mixing coefficient of the blurred image corresponding to the extended image may be -0.5.
  • adjusting the contrast of the clear image includes: adjusting the gray value of each pixel of the clear image according to the gray mean value of the clear image. Therefore, by adjusting the contrast of the clear image, a second binarized image with more obvious black and white contrast can be obtained.
  • the gray value of each pixel of a clear image can be adjusted by the following formula:
  • f'(i,j) is the gray value of the pixel point at (i,j) of the second binarized image, is the average value of the gray value of each pixel point in the clear image, f(i, j) is the gray value of the pixel point at (i, j) of the clear image, and t is the intensity value.
  • the intensity value may be from 0.1 to 0.5, for example, the intensity value may be 0.2. In practical applications, the intensity value can be selected according to the final ink saving effect to be achieved.
  • FIG. 8 is a schematic diagram of a composite image provided by an embodiment of the present disclosure.
  • step S14 the second binarized image and the first binarized image are synthesized according to the positions of the plurality of circumscribed contour pixels in the pixel circumscribed contour image to obtain a synthesized image.
  • combining the second binarized image and the first binarized image to obtain a composite image means combining the pixels corresponding to the positions of the plurality of circumscribed contour pixels in the first binarized image.
  • the gray-scale value is replaced with the gray-scale value of the pixels corresponding to the positions of the multiple circumscribed contour pixels in the second binarized image, that is, all the pixels corresponding to the positions of the multiple circumscribed contour pixels in the first binarized image Replace with better-performing pixels.
  • FIG. 8 is an image obtained by synthesizing the first binarized image shown in FIG. 4 and the second binarized image shown in FIG. 7 .
  • the composite image is a binarized image.
  • step S14 includes: obtaining the positions of a plurality of circumscribed contour pixels in the pixel circumscribed contour image; Two binarized pixels; according to the pixel correspondence between the second binarized image and the first binarized image, multiple target second binarized pixels in the second binarized image are synthesized into the first binarized image respectively. The same location in the image to get a composite image of the original image.
  • the size of the second binarized image, the size of the first binarized image, and the size of the composite image may all be the same.
  • All the second binarized pixels in the second binarized image are arranged in n rows and m columns, all the first binarized pixels in the first binarized image are arranged in n rows and m columns, and all the composite images in the composite image are arranged in n rows and m columns.
  • the pixels are arranged in n rows and m columns, and all the pixels in the pixel circumscribed contour image are arranged in n rows and m columns.
  • the number of all the second binarized pixels in the second binarized image is n*m
  • the number of all the first binarized pixels in the first binarized image is n*m
  • the composite image The number of all synthesized pixels in is n*m
  • the number of all pixels in the pixel circumscribed contour image is n*m.
  • the multiple target second binarized pixels are in one-to-one correspondence with the multiple circumscribed contour pixels, and the positions of the target second binarized pixels are the same as the positions of the circumscribed contour pixels corresponding to the target second binarized pixels. For example, if in the pixel circumscribed contour image, the pixel located in the i-th row and the j-th column is the circumscribed contour pixel, correspondingly, in the second binarized image, the second binarized pixel located in the i-th row and the j-th column The second binarized pixel for the target.
  • step S14 according to the pixel correspondence between the second binarized image and the first binarized image, multiple target second binarized pixels in the second binarized image are synthesized into the first binarized image respectively.
  • the same position in the image includes: determining the qth target second binarization pixel in the plurality of target second binarization pixels, wherein the qth target second binarization pixel is located in the second binarized image The i-th row and the j-th column of ; determine the q-th target first binarized pixel located in the i-th row and the j-th column in the first binarized image; The value is replaced with the grayscale value of the qth target second binarized pixel to obtain the qth target composite pixel located at the ith row and the jth column in the composite image.
  • n, m, q, i, and j are all positive integers, and i is less than or equal to n, j is less than or equal to m, and q is less than or equal to the number of multiple target second binarized pixels.
  • the first binarized pixels that do not correspond to the plurality of target second binarized pixels are directly used as the synthesized pixels in the synthesized image.
  • the first binarized pixel located in the p1th row and the p2th column does not correspond to any of the multiple target second binarization pixels, that is, the second binarized image in the second binarized image is located in
  • the second binarized pixel at row p1 and column p2 is not the target second binarized pixel, then the first binarized pixel located at row p1 and column p2 in the first binarized image is used as the composite image.
  • the grayscale value of the synthesized pixel located at row p1 and column p2, that is, the grayscale value of the first binarized pixel located at row p1 and column p2 is the same as the grayscale value of the synthesized pixel located at row p1 and column p2.
  • the pixels in the second binarized image are referred to as second binarized pixels
  • the pixels in the first binarized image are referred to as first binarized pixels
  • the composite image The pixels in the image are called composite pixels.
  • "Second binarized pixels”, “first binarized pixels”, “synthetic pixels”, etc. are only used to distinguish pixels located in different images, and do not mean that these pixels are in different images. structure, properties, etc.
  • the target second binarized pixel represents a pixel corresponding to the circumscribed contour pixel in the second binarized image
  • the target first binarized pixel represents a pixel corresponding to the target second binarized pixel in the first binarized image.
  • the corresponding pixel, the target composite pixel represents a pixel in the composite image corresponding to the target second binarized pixel.
  • FIG. 9 is a schematic block diagram of an image processing apparatus provided by at least one embodiment of the present disclosure.
  • the image processing apparatus 900 includes: an acquisition module 901 , a first binarization module 902 , a processing module 903 , a second binarization module 904 and a synthesis module 905 .
  • the acquisition module 901 is used to acquire the original image.
  • the original image includes at least one object.
  • the first binarization module 902 is configured to process the original image through the first binarization model to obtain a first binarized image of the original image.
  • the processing module 903 is configured to process the first binarized image to obtain a pixel circumscribed contour image.
  • the pixel circumscribed contour image includes a plurality of circumscribed contour pixels, and the pixels in the area surrounded by the plurality of circumscribed contour pixels are pixels corresponding to at least some objects in the at least one object.
  • the second binarization module 904 is configured to process the original image through the second binarization model to obtain a second binarized image.
  • the synthesizing module 905 is configured to synthesize the second binarized image and the first binarized image according to the positions of a plurality of circumscribed contour pixels in the pixel circumscribed contour image to obtain a synthesized image of the original image.
  • the composite image is a binarized image.
  • the first binarization module 902 when the first binarization module 902 performs the operation of processing the original image through the first binarization model to obtain the first binarized image of the original image, the first binarization module 902 performs The following operations: compress the original image to obtain the input image, wherein the size of the input image is smaller than the size of the original image; process the input image through the first binarization model to obtain the first binarization of the original image image.
  • the first binarization module 902 when the first binarization module 902 performs the operation of processing the input image through the first binarization model to obtain the first binarized image of the original image, the first binarization module 902 performs The following operations are as follows: processing the input image through the first binarization model to obtain a binarized predicted image of the input image; and restoring the size of the binarized predicted image to obtain the first binarized image.
  • the size of the first binarized image is the same as the size of the original image.
  • the processing module 903 when the processing module 903 performs an operation of processing the first binarized image to obtain a pixel circumscribed contour image, the processing module 903 performs the following operations: performing blurring processing on the first binarized image to obtain Blurred image; XOR processing is performed on the blurred image and the first binarized image to obtain a pixel circumscribed contour image.
  • the synthesis module 905 performs synthesis of the second binarized image and the first binarized image according to the positions of a plurality of circumscribed contour pixels in the pixel circumscribed contour image to obtain a composite image of the original image.
  • the synthesizing module 905 performs the following operations: obtaining the positions of multiple circumscribed contour pixels in the pixel circumscribed contour image; Two binarized pixels; according to the pixel correspondence between the second binarized image and the first binarized image, multiple target second binarized pixels in the second binarized image are synthesized into the first binarized image respectively. The same location in the image to get a composite image of the original image.
  • All the second binarized pixels in the second binarized image are arranged in n rows and m columns, all the first binarized pixels in the first binarized image are arranged in n rows and m columns, and all the composite images in the composite image are arranged in n rows and m columns.
  • the pixels are arranged in n rows and m columns.
  • the synthesizing module 905 performs, according to the pixel correspondence between the second binarized image and the first binarized image, respectively synthesizing a plurality of target second binarized pixels in the second binarized image into During the operation at the same position in the first binarized image, the synthesizing module 905 includes performing the following operations: determining the qth target second binarization pixel among the plurality of target second binarization pixels, wherein the qth target second binarization pixel is The target second binarization pixel is located in the ith row and the jth column of the second binarized image; determine the qth target first binarization pixel located in the ith row and the jth column in the first binarized image; Replace the grayscale value of the first binarized pixel of the qth target with the grayscale value of the second binarized pixel of the qth target to obtain the composite image of the qth target located at the ith row and the jth column of the composite
  • the grayscale value of the qth target synthetic pixel is the grayscale value of the qth target second binarized pixel, n, m, q, i, and j are all positive integers, and i is less than or equal to n, and j is less than or equal to m , q is less than or equal to the number of multiple target second binarized pixels.
  • the second binarization module 904 includes a grayscale module, an intermediate binarization module, a filter module, a determination module, an expansion module, and a sharpening module.
  • the grayscale module is used to perform grayscale processing on the original image to obtain a grayscale image.
  • the intermediate binarization module is used to process the grayscale image according to the first threshold to obtain an intermediate binarized image.
  • the filtering module is used for taking the intermediate binarized image as a guide image, and performing guide filtering processing on the grayscale image to obtain a filtered image.
  • the determining module is configured to determine high-value pixels in the filtered image according to the second threshold, wherein the gray value of the high-value pixels is greater than the second threshold.
  • the expansion module is used for expanding the gray value of the high-value pixel points according to the preset expansion coefficient to obtain an expanded image.
  • the sharpening module is used for sharpening the expanded image to obtain a clear image, and adjusting the contrast of the clear image to obtain a second binarized image.
  • acquisition module 901, first binarization module 902, processing module 903, second binarization module 904, and/or synthesis module 905 include code and programs stored in memory; the code and programs can be executed by the processor to Some or all of the functions of the acquisition module 901 , the first binarization module 902 , the processing module 903 , the second binarization module 904 and/or the synthesis module 905 as described above are implemented.
  • the acquisition module 901 , the first binarization module 902 , the processing module 903 , the second binarization module 904 and/or the synthesis module 905 may be dedicated hardware devices for implementing the acquisition module 901 , the first Some or all of the functions of the binarization module 902 , the processing module 903 , the second binarization module 904 and/or the synthesis module 905 .
  • the acquisition module 901, the first binarization module 902, the processing module 903, the second binarization module 904 and/or the synthesis module 905 may be one circuit board or a combination of multiple circuit boards for implementing the above-mentioned function.
  • the one circuit board or the combination of multiple circuit boards may include: (1) one or more processors; (2) one or more non-transitory memories connected to the processors; and (3) The firmware stored in the memory executable by the processor.
  • the acquisition module 901 is used to implement step S10 shown in FIG. 1
  • the first binarization module 902 is used to implement step S11 shown in FIG. 1
  • the processing module 903 is used to implement step S12 shown in FIG. 1
  • the second binarization module 904 is used to implement step S13 shown in FIG. 1
  • the synthesis module 905 is used to implement step S14 shown in FIG. 1 . Therefore, for the specific description of the acquisition module 901, reference may be made to the relevant description of step S10 shown in FIG. 1 in the above-mentioned embodiment of the image processing method, and for the specific description of the first binarization module 902, reference may be made to the above-mentioned embodiment of the image processing method.
  • step S11 shown in FIG. 1 for the specific description of the processing module 903 , please refer to the relevant description of step S12 shown in FIG. 1 in the embodiment of the above image processing method, and for the specific description of the second binarization module 904 Reference may be made to the relevant description of step S13 shown in FIG. 1 in the embodiment of the above image processing method, and the specific description of the synthesis module 905 may refer to the relevant description of step S14 shown in FIG. 1 in the embodiment of the above image processing method.
  • the image processing apparatus can achieve technical effects similar to those of the aforementioned image processing method, which will not be repeated here.
  • FIG. 10 is a schematic block diagram of an electronic device provided by at least one embodiment of the present disclosure.
  • the electronic device includes a processor 1001 , a communication interface 1002 , a memory 1003 and a communication bus 1004 .
  • the processor 1001, the communication interface 1002, and the memory 1003 communicate with each other through the communication bus 1004, and the components such as the processor 1001, the communication interface 1002, and the memory 1003 can also communicate through a network connection.
  • the present disclosure does not limit the type and function of the network.
  • Memory 1003 is used for non-transitory storage of computer readable instructions.
  • the processor 1001 When the processor 1001 is configured to execute computer-readable instructions, the computer-readable instructions are executed by the processor 1001 to implement the image processing method according to any of the above embodiments.
  • the processor 1001 For the specific implementation of each step of the image processing method and related explanation contents, reference may be made to the above-mentioned embodiments of the image processing method, which will not be repeated here.
  • the implementation manner of the processor 1001 executing the program stored in the memory 1003 to realize the image processing method is the same as the implementation manner mentioned in the embodiment part of the foregoing image processing method, and will not be repeated here.
  • the communication bus 1004 may be a Peripheral Component Interconnect Standard (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, or the like.
  • PCI Peripheral Component Interconnect Standard
  • EISA Extended Industry Standard Architecture
  • the communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.
  • the communication interface 1002 is used to implement communication between the electronic device and other devices.
  • the processor 1001 and the memory 1003 may be provided on the server side (or cloud).
  • the processor 1001 may control other components in the electronic device to perform desired functions.
  • the processor 1001 can be a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable Logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • the central processing unit (CPU) can be an X86 or an ARM architecture or the like.
  • Memory 1003 may include any combination of one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory.
  • Volatile memory may include, for example, random access memory (RAM) and/or cache memory, among others.
  • Non-volatile memory may include, for example, read only memory (ROM), hard disk, erasable programmable read only memory (EPROM), portable compact disk read only memory (CD-ROM), USB memory, flash memory, and the like.
  • ROM read only memory
  • EPROM erasable programmable read only memory
  • CD-ROM portable compact disk read only memory
  • USB memory flash memory
  • flash memory flash memory
  • FIG. 11 is a schematic diagram of a non-transitory computer-readable storage medium provided by at least one embodiment of the present disclosure.
  • one or more computer-readable instructions 1101 may be non-transitory stored on storage medium 1100 .
  • the computer readable instructions 1101 may perform one or more steps of the image processing method according to the above when executed by a processor.
  • the storage medium 1100 can be applied to the above-mentioned electronic device and/or the image processing apparatus 900 .
  • the storage medium 1100 may include the memory 1003 in the electronic device.
  • FIG. 12 shows a schematic diagram of a hardware environment provided by at least one embodiment of the present disclosure.
  • the electronic device provided by the present disclosure can be applied to the Internet system.
  • the functions of the image processing apparatus and/or electronic device involved in the present disclosure can be realized by using the computer system provided in FIG. 12 .
  • Such computer systems may include personal computers, notebook computers, tablet computers, cell phones, personal digital assistants, smart glasses, smart watches, smart rings, smart helmets, and any smart portable or wearable device.
  • the specific system in this embodiment illustrates a hardware platform including a user interface using functional block diagrams.
  • Such computer equipment may be a general purpose computer equipment or a special purpose computer equipment. Both computer devices can be used to implement the image processing apparatus and/or electronic device in this embodiment.
  • the computer system may include any component that implements the information required to implement the image processing currently described.
  • a computer system can be implemented by a computer device through its hardware devices, software programs, firmware, and combinations thereof.
  • FIG. 12 only one computer device is drawn in FIG. 12, but the related computer functions described in this embodiment to realize the information required for image processing can be implemented in a distributed manner by a group of similar platforms, Distribute the processing load of a computer
  • the computer system may include a communication port 250, which is connected to a network for realizing data communication.
  • the computer system may send and receive information and data through the communication port 250, that is, the communication port 250 may enable the computer system to communicate with Other electronic devices communicate wirelessly or by wire to exchange data.
  • the computer system may also include a processor group 220 (ie, the processors described above) for executing program instructions.
  • the processor group 220 may consist of at least one processor (eg, a CPU).
  • the computer system may include an internal communication bus 210 .
  • the computer system may include various forms of program storage units and data storage units (ie, the memories or storage media described above), such as hard disk 270, read only memory (ROM) 230, random access memory (RAM) 240, capable of storing Various data files used for computer processing and/or communication, and possibly program instructions executed by the processor group 220 .
  • the computer system may also include an input/output component 260 for enabling input/output data flow between the computer system and other components (eg, user interface 280, etc.).
  • input devices including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, a liquid crystal display (LCD), speakers, vibrators, etc. output device; including storage devices such as tapes, hard disks, etc.; and a communication interface.
  • input devices including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.
  • LCD liquid crystal display
  • speakers vibrators
  • storage devices such as tapes, hard disks, etc.
  • communication interface such as tapes, hard disks, etc.
  • Figure 12 shows a computer system with various devices, it should be understood that the computer system is not required to have all of the devices shown, and may instead have more or fewer devices.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Processing (AREA)

Abstract

L'invention concerne un procédé de traitement d'image, un appareil de traitement d'image, un dispositif électronique et un support de stockage lisible par ordinateur non transitoire. Le procédé de traitement d'image comprend les étapes consistant à : obtenir une image d'origine, l'image d'origine comprenant au moins un objet ; traiter l'image d'origine par un premier modèle de binarisation pour obtenir une première image binarisée de l'image d'origine ; traiter la première image binarisée afin d'obtenir une image de contour circonscrit de pixel, l'image de contour circonscrit de pixel comprenant une pluralité de pixels de contour circonscrits, et des pixels à l'intérieur d'une zone entourée par la pluralité de pixels de contour circonscrits étant des pixels correspondant à au moins une partie des objets dans le ou les objets ; traiter l'image d'origine par un second modèle de binarisation pour obtenir une seconde image binarisée ; et en fonction des positions de la pluralité de pixels de contour circonscrits dans l'image de contour circonscrit de pixel, synthétiser la seconde image binarisée et la première image binarisée afin d'obtenir une image synthétisée, l'image composite étant une image binarisée.
PCT/CN2021/102912 2020-07-03 2021-06-29 Procédé de traitement d'image, appareil de traitement d'image, dispositif électronique et support de stockage WO2022002002A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010636558.4A CN111767924B (zh) 2020-07-03 2020-07-03 图像处理方法、图像处理装置、电子设备、存储介质
CN202010636558.4 2020-07-03

Publications (1)

Publication Number Publication Date
WO2022002002A1 true WO2022002002A1 (fr) 2022-01-06

Family

ID=72724061

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/102912 WO2022002002A1 (fr) 2020-07-03 2021-06-29 Procédé de traitement d'image, appareil de traitement d'image, dispositif électronique et support de stockage

Country Status (2)

Country Link
CN (1) CN111767924B (fr)
WO (1) WO2022002002A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111767924B (zh) * 2020-07-03 2024-01-26 杭州睿琪软件有限公司 图像处理方法、图像处理装置、电子设备、存储介质
CN115100228B (zh) * 2022-07-25 2022-12-20 江西现代职业技术学院 图像处理方法、系统、可读存储介质及计算机设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101443897A (zh) * 2006-05-16 2009-05-27 东京毅力科创株式会社 图像二值化方法、图像处理装置及计算机程序
CN109785369A (zh) * 2017-11-10 2019-05-21 中国移动通信有限公司研究院 一种虚拟现实人像采集方法及装置
CN110008954A (zh) * 2019-03-29 2019-07-12 重庆大学 一种基于多阈值融合的复杂背景文本图像提取方法及系统
US20200082208A1 (en) * 2018-09-06 2020-03-12 International Business Machines Corporation Image binarization using mean restrain
CN111325657A (zh) * 2020-02-18 2020-06-23 北京奇艺世纪科技有限公司 图像处理方法、装置、电子设备和计算机可读存储介质
CN111767924A (zh) * 2020-07-03 2020-10-13 杭州睿琪软件有限公司 图像处理方法、图像处理装置、电子设备、存储介质

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107563977A (zh) * 2017-08-28 2018-01-09 维沃移动通信有限公司 一种图像处理方法、移动终端及计算机可读存储介质
CN109754379A (zh) * 2018-12-29 2019-05-14 北京金山安全软件有限公司 一种图像处理方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101443897A (zh) * 2006-05-16 2009-05-27 东京毅力科创株式会社 图像二值化方法、图像处理装置及计算机程序
CN109785369A (zh) * 2017-11-10 2019-05-21 中国移动通信有限公司研究院 一种虚拟现实人像采集方法及装置
US20200082208A1 (en) * 2018-09-06 2020-03-12 International Business Machines Corporation Image binarization using mean restrain
CN110008954A (zh) * 2019-03-29 2019-07-12 重庆大学 一种基于多阈值融合的复杂背景文本图像提取方法及系统
CN111325657A (zh) * 2020-02-18 2020-06-23 北京奇艺世纪科技有限公司 图像处理方法、装置、电子设备和计算机可读存储介质
CN111767924A (zh) * 2020-07-03 2020-10-13 杭州睿琪软件有限公司 图像处理方法、图像处理装置、电子设备、存储介质

Also Published As

Publication number Publication date
CN111767924A (zh) 2020-10-13
CN111767924B (zh) 2024-01-26

Similar Documents

Publication Publication Date Title
US20230222631A1 (en) Method and device for removing handwritten content from text image, and storage medium
WO2021233266A1 (fr) Procédé et appareil de détection de bord et dispositif électronique et support de stockage
US20190304066A1 (en) Synthesis method of chinese printed character images and device thereof
WO2021147631A1 (fr) Procédé et dispositif d'élimination de contenu manuscrit et support d'enregistrement
CN113486828B (zh) 图像处理方法、装置、设备和存储介质
WO2022002002A1 (fr) Procédé de traitement d'image, appareil de traitement d'image, dispositif électronique et support de stockage
CN109117846B (zh) 一种图像处理方法、装置、电子设备和计算机可读介质
US20150070373A1 (en) Clarification of Zoomed Text Embedded in Images
US20210279509A1 (en) Method and System For Processing Images Using Cross-Stage Skip Connections
WO2021146937A1 (fr) Procédé de reconnaissance de caractères, dispositif de reconnaissance de caractères et support de stockage
Demilew et al. Ancient Geez script recognition using deep learning
KR20130066819A (ko) 촬영 이미지 기반의 문자 인식 장치 및 방법
CN110717497A (zh) 图像相似度匹配方法、装置及计算机可读存储介质
CN113436222A (zh) 图像处理方法、图像处理装置、电子设备及存储介质
US11985287B2 (en) Image processing method, image processing device, electronic apparatus and storage medium
CN111047618B (zh) 基于多尺度的无参考屏幕内容图像质量评估方法
US20160259974A1 (en) Selective, user-mediated content recognition using mobile devices
RU2633182C1 (ru) Определение направления строк текста
US7676089B2 (en) Document layout analysis with control of non-character area
WO2023130966A1 (fr) Procédé de traitement d'image, appareil de traitement d'image, dispositif électronique et support de stockage
WO2022247702A1 (fr) Procédé et appareil de traitement d'image, dispositif électronique et support de stockage
CN114170589A (zh) 一种基于nas的岩石岩性识别方法、终端设备及存储介质
Panchal et al. An investigation on feature and text extraction from images using image recognition in Android
CN106934814B (zh) 一种基于图像的背景信息识别方法及装置
CN113223025B (zh) 图像处理方法及装置、神经网络的训练方法及装置

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21834554

Country of ref document: EP

Kind code of ref document: A1