WO2021147387A1 - 屏幕划痕碎裂检测方法及设备 - Google Patents

屏幕划痕碎裂检测方法及设备 Download PDF

Info

Publication number
WO2021147387A1
WO2021147387A1 PCT/CN2020/120892 CN2020120892W WO2021147387A1 WO 2021147387 A1 WO2021147387 A1 WO 2021147387A1 CN 2020120892 W CN2020120892 W CN 2020120892W WO 2021147387 A1 WO2021147387 A1 WO 2021147387A1
Authority
WO
WIPO (PCT)
Prior art keywords
screen image
category
image
screen
target candidate
Prior art date
Application number
PCT/CN2020/120892
Other languages
English (en)
French (fr)
Inventor
陈敏
Original Assignee
上海万物新生环保科技集团有限公司
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 上海万物新生环保科技集团有限公司 filed Critical 上海万物新生环保科技集团有限公司
Publication of WO2021147387A1 publication Critical patent/WO2021147387A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the invention relates to the field of computers, in particular to a method and equipment for detecting screen scratches and chipping.
  • An object of the present invention is to provide a method and equipment for detecting screen scratches and chipping.
  • a method for detecting screen scratches and chipping including:
  • Control the screen to display a full-screen yellow image below the preset exposure value, and take a yellow screen image based on the outline position of the screen;
  • Control the screen to display a full-screen black image higher than the preset exposure value, and take a black screen image based on the outline position of the screen;
  • target candidate frames in the yellow screen image and the black screen image are obtained in which the target categories are the scratch pattern category and the broken crack category.
  • the convolutional neural network is a resnext101 convolutional neural network.
  • the yellow screen image and the black screen image are obtained, and the target categories are the target candidate frames of the scratch pattern category and the broken crack category.
  • the corresponding multi-layer feature layers of different scales corresponding to the yellow screen image are obtained; based on the image features corresponding to the black screen image and the corresponding black screen image Image features, and using the FPN method to obtain multi-layer feature layers of different scales corresponding to the corresponding black screen image;
  • the target candidate frame in the yellow screen image is extracted through the RPN network on the multi-layer feature layers of different scales corresponding to the yellow screen image, and each target candidate frame in the yellow screen image is preset to have scratches The probability value of grains and cracks;
  • the target candidate frame in the black screen image is extracted on the multi-layer feature layers of different scales corresponding to the black screen image through the RPN network, and the target candidate frame in the black screen image is preset The probability value of scratches and cracks in each target candidate frame;
  • the first preset number of target candidate frames in the yellow screen image are input into the classification neural network, and each target candidate frame in the first preset number of target candidate frames in the yellow screen image corresponding to the output is obtained
  • the probability values belonging to the background category, the scratch pattern category and the broken crack category respectively input the previously preset number of target candidate frames in the black screen image into the classification neural network, and obtain the corresponding output black screen image
  • the initial category is determined as the target category of the target candidate frame
  • the output determines the target candidate frame with the target category as scratch pattern category and broken crack category.
  • outputting the target candidate frames whose target categories are determined to be the scratch pattern category and the broken crack category include:
  • the target candidate frames in the yellow screen image with the positions of the determined target categories overlapped are sorted in descending order based on the probability value to obtain the first sorting queue, and the target candidate frame with the highest probability value in the first sorting queue is used as the first reference Candidate frame, if the overlapping area of each target candidate frame in the subsequent queues in the first sorting queue and the first reference candidate frame exceeds the threshold of the area of the first reference candidate frame of the preset ratio, then The target candidate frame and its corresponding target category are deleted; the target candidate frames in the black screen image whose positions where the target category is determined overlap are arranged in descending order based on the probability value to obtain a second sorting queue, and the second sorting queue is placed in the second sorting queue.
  • the target candidate frame with the highest probability value is used as the second reference candidate frame. If the overlapping area of each subsequent target candidate frame in the second sorting queue and the second reference candidate frame exceeds the preset ratio of the second reference candidate frame The threshold of the area of, delete the target candidate frame and its corresponding target category;
  • the output determines the target candidate frame with the target category as scratch pattern category and broken crack category.
  • the classification neural network is a fully connected layer classification neural network.
  • determining the outline position of the screen includes:
  • the boundary of the white background picture is recognized from the photo, and the boundary is taken as the position of the outline of the screen.
  • identifying the boundary of the white background picture from the photo and using the boundary as the position of the outline of the screen includes:
  • the area where the remaining pixel points are continuous and the fullness s is greater than the preset fullness threshold T3 is taken as the boundary of the white background picture, and the boundary is taken as the position of the outline of the screen.
  • a screen scratch and chipping detection device wherein the device includes:
  • Positioning device used to determine the contour position of the screen
  • the display and shooting device is used to control the screen to display a full-screen yellow image below the preset exposure value, and shoot the yellow screen image based on the outline position of the screen; control the screen to display a full-screen black image higher than the preset exposure value, based on The outline position of the screen, taking a black screen image;
  • a feature extraction device for inputting the yellow screen image into a convolutional neural network, and extracting image features corresponding to the yellow screen image; inputting the black screen image into the convolutional neural network, and extracting the corresponding black screen image Image characteristics;
  • An identification device is used to obtain target candidate frames in the yellow screen image and the black screen image whose target categories are the scratch pattern category and the broken crack category based on the image features corresponding to the yellow screen image and the black screen image respectively.
  • the convolutional neural network is a resnext101 convolutional neural network.
  • the identification device is used for:
  • the corresponding multi-layer feature layers of different scales corresponding to the yellow screen image are obtained; based on the image features corresponding to the black screen image and the corresponding black screen image Image features, and using the FPN method to obtain multi-layer feature layers of different scales corresponding to the corresponding black screen image;
  • the target candidate frame in the yellow screen image is extracted through the RPN network on the multi-layer feature layers of different scales corresponding to the yellow screen image, and each target candidate frame in the yellow screen image is preset to have scratches The probability value of grains and cracks;
  • the target candidate frame in the black screen image is extracted on the multi-layer feature layers of different scales corresponding to the black screen image through the RPN network, and the target candidate frame in the black screen image is preset The probability value of scratches and cracks in each target candidate frame;
  • the first preset number of target candidate frames in the yellow screen image are input into the classification neural network, and each target candidate frame in the first preset number of target candidate frames in the yellow screen image corresponding to the output is obtained
  • the probability values belonging to the background category, the scratch pattern category and the broken crack category respectively input the previously preset number of target candidate frames in the black screen image into the classification neural network, and obtain the corresponding output black screen image
  • the initial category is determined as the target category of the target candidate frame
  • the output determines the target candidate frame with the target category as scratch pattern category and broken crack category.
  • the identification device is used for:
  • the target candidate frames in the yellow screen image with the positions of the determined target categories overlapped are sorted in descending order based on the probability value to obtain the first sorting queue, and the target candidate frame with the highest probability value in the first sorting queue is used as the first reference Candidate frame, if the overlapping area of each target candidate frame in the subsequent queues in the first sorting queue and the first reference candidate frame exceeds the threshold of the area of the first reference candidate frame of the preset ratio, then The target candidate frame and its corresponding target category are deleted; the target candidate frames in the black screen image whose positions where the target category is determined overlap are arranged in descending order based on the probability value to obtain a second sorting queue, and the second sorting queue is placed in the second sorting queue.
  • the target candidate frame with the highest probability value is used as the second reference candidate frame. If the overlapping area of each subsequent target candidate frame in the second sorting queue and the second reference candidate frame exceeds the preset ratio of the second reference candidate frame The threshold of the area of, delete the target candidate frame and its corresponding target category;
  • the output determines the target candidate frame with the target category as scratch pattern category and broken crack category.
  • the classification neural network is a fully connected layer classification neural network.
  • the positioning device includes:
  • Display module used to display the bright screen of the screen as a white background picture
  • the recognition module is configured to recognize the border of the white background picture from the photo, and use the border as the position of the outline of the screen.
  • the recognition module is configured to convert the photo into a grayscale image; specify a preset pixel threshold T1 to segment the grayscale image, wherein the photo exceeds the preset Set the pixel value of the pixel of the pixel threshold T1 to 255, and set the pixel value of the pixel in the photo that does not exceed the preset pixel threshold T1 to 0; obtain the pixel value of 255 in the grayscale image
  • a computing-based device which includes:
  • a memory arranged to store computer-executable instructions which, when executed, cause the processor to:
  • Control the screen to display a full-screen yellow image below the preset exposure value, and take a yellow screen image based on the outline position of the screen;
  • Control the screen to display a full-screen black image higher than the preset exposure value, and take a black screen image based on the outline position of the screen;
  • target candidate frames in the yellow screen image and the black screen image are obtained in which the target categories are the scratch pattern category and the broken crack category.
  • a computer-readable storage medium having computer-executable instructions stored thereon, wherein, when the computer-executable instructions are executed by a processor, the processor:
  • Control the screen to display a full-screen yellow image below the preset exposure value, and take a yellow screen image based on the outline position of the screen;
  • Control the screen to display a full-screen black image higher than the preset exposure value, and take a black screen image based on the outline position of the screen;
  • target candidate frames in the yellow screen image and the black screen image are obtained in which the target categories are the scratch pattern category and the broken crack category.
  • the present invention obtains that the target categories in the yellow screen image and the black screen image are scratch marks and cracks based on the corresponding image features of the yellow screen image and the black screen image.
  • the target candidate frame can accurately identify scratches or cracks on the screen of mobile phones and other devices, and can improve the efficiency of mobile phone and other smart devices such as valuation and recycling.
  • FIG. 1 shows a flowchart of a method for detecting screen scratches and chipping according to an embodiment of the present invention.
  • the terminal, the equipment of the service network, and the trusted party all include one or more processors (CPU), input/output interfaces, network interfaces, and memory.
  • processors CPU
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • Memory may include non-permanent memory in computer-readable media, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer readable media.
  • RAM random access memory
  • ROM read-only memory
  • flash RAM flash memory
  • Computer-readable media includes permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology.
  • the information can be computer readable instructions, data structures, program modules, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, Magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices.
  • computer-readable media does not include non-transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.
  • the present invention provides a method for detecting screen scratches and chipping.
  • the method includes:
  • Step S0 determine the outline position of the screen
  • Step S1 controlling the screen to display a full-screen yellow image lower than the preset exposure value, and taking a yellow screen image based on the outline position of the screen;
  • Step S2 controlling the screen to display a full-screen black image higher than the preset exposure value, and taking a black screen image based on the outline position of the screen;
  • high-exposure pictures are conducive to shooting dark screen surface textures, but for bright screen surface textures it is easy to cause overexposure problems, so low-exposure pictures need to be used for auxiliary detection;
  • Step S3 input the yellow screen image into the convolutional neural network, and extract the image features corresponding to the yellow screen image; input the black screen image into the convolutional neural network, and extract the image features corresponding to the black screen image ;
  • the convolutional neural network may be a resnext101 convolutional neural network to extract accurate image features
  • Step S4 based on the image features corresponding to the yellow screen image and the black screen image, respectively, obtain target candidate frames in the yellow screen image and the black screen image whose target categories are the scratch pattern category and the broken crack category.
  • the present invention obtains target candidate frames in the yellow screen image and the black screen image whose target categories are the scratch pattern category and the broken crack category based on the image features corresponding to the yellow screen image and the black screen image, respectively. It can accurately identify scratches or cracks on the screen of mobile phones and other devices, and can improve the efficiency of valuation and recycling of smart devices such as mobile phones.
  • step S4 based on the image characteristics corresponding to the yellow screen image and the black screen image, respectively, the yellow screen image and the black screen image are obtained, and the target category is scratch
  • the target candidate frames of the mark type and the crack type include:
  • Step S41 Based on the image features corresponding to the yellow screen image, and through the FPN (feature pyramid networks) method, obtain the corresponding multi-layer feature layers of different scales corresponding to the yellow screen image; based on the corresponding black screen image Image features and image features corresponding to the black screen image, and using the FPN method to obtain the corresponding multi-layer feature layers of different scales corresponding to the black screen image;
  • FPN feature pyramid networks
  • step S42 the target candidate frame in the yellow screen image is extracted on the multi-layer feature layers of different scales corresponding to the yellow screen image through the RPN (Region Proposal Network) network, and the target candidate frame in the yellow screen image is preset
  • RPN Registered Proposal Network
  • Each target candidate frame has the probability value of scratches and broken cracks
  • the target candidate frame in the black screen image is extracted through the RPN network on the multi-layer feature layers of different scales corresponding to the black screen image, and the prediction is performed It is assumed that each target candidate frame in the black screen image has a probability value of scratches and cracks;
  • Step S43 selecting the first preset number of target candidate frames in the yellow screen image with a larger probability value; selecting the first preset number of target candidate frames in the black screen image with a larger probability value;
  • the first 1000 target candidate frames in the yellow screen image with a larger probability value may be selected; the first 1000 target candidate frames in the black screen image with a larger probability value may be selected;
  • Step S44 Input the previously preset number of target candidate frames in the yellow screen image into the classification neural network, and obtain each corresponding output of the first preset number of target candidate frames in the yellow screen image
  • the target candidate frames respectively belong to the probability values of the background category, the scratch pattern category and the broken crack category
  • the probability value that each target candidate frame in the previously preset number of target candidate frames in the black screen image belongs to the background category, the scratch pattern category and the broken crack category respectively;
  • the classification neural network may be a fully connected layer classification neural network to obtain reliable classification
  • Step S45 Determine the corresponding category with a larger probability value of each target candidate frame as the initial category of the target candidate frame;
  • the neural network outputs a target candidate frame a with a background category probability value of 0.2, a scratch pattern category has a probability value of 0.3, and a broken crack category has a probability value of 0.5, then the target candidate frame a
  • the initial category is the crack category
  • the neural network outputs the probability value of the background category of a certain target candidate frame b as 0.1, the probability value of the scratch pattern category is 0.2, and the probability value of the broken crack category is 0.7, then the initial value of the target candidate frame b The category is broken and cracked;
  • Step S46 if it is determined that the probability value of the initial category of the target candidate frame of the initial category is greater than the preset probability threshold, then the initial category is determined as the target category of the target candidate frame;
  • the preset probability threshold is 0.6
  • the neural network outputs that the initial category of a certain target candidate frame a is the broken crack category, and the probability value of the broken crack category is 0.5. Since the predetermined probability threshold of 0.6 is not exceeded, the initial category of the broken crack category of the target candidate frame a The category cannot be used as the target category;
  • the neural network outputs the initial category of a certain target candidate frame b as the broken crack category, and the probability value of the broken crack category is 0.7. Since the predetermined probability threshold of 0.6 is exceeded, the broken crack category of the target candidate frame b The initial category of can be used as the target category;
  • Step S47 outputting a target candidate frame whose target categories are determined to be the scratch pattern category and the broken crack category.
  • the initial category of the target candidate frame is determined, and then the target candidate frame of the determined target category is filtered from the target candidate frame of the determined initial category, which can further reliably and accurately identify the screen of the mobile phone and other devices. Scratches or cracks.
  • step S47 outputting target candidate frames whose target categories are determined to be the scratch pattern category and the broken crack category, including:
  • Step S471 Arrange the target candidate frames in the yellow screen image that overlap the target categories in descending order based on the probability value to obtain a first sorting queue, and use the target candidate frame with the highest probability value in the first sorting queue as The first reference candidate frame, if the overlapping area of each target candidate frame in the subsequent queues in the first sorting queue and the first reference candidate frame exceeds the threshold value of the area of the first reference candidate frame of the preset ratio , The target candidate frame and its corresponding target category are deleted; the target candidate frames in the black screen image where the target category is determined overlapped are arranged in descending order based on the probability value to obtain a second sorting queue, and the second sorting queue is obtained.
  • the target candidate frame with the highest probability value in the sorting queue is used as the second reference candidate frame. If the overlapping area of each subsequent target candidate frame in the second sorting queue and the second reference candidate frame exceeds the preset ratio of the second The threshold of the area of the reference candidate frame, the target candidate frame and its corresponding target category are deleted;
  • step S472 the target candidate frames whose target categories are determined to be the scratch pattern category and the broken crack category are output.
  • the preset ratio threshold may be 0.7.
  • step S0 determining the outline position of the screen, includes:
  • Step S01 display the bright screen of the screen as a white background picture
  • the screen may be a terminal device with a display screen, such as a mobile phone or a PAD;
  • Step S02 taking a picture of the screen including the white background picture
  • Step S03 Identify the border of the white background picture from the photo, and use the border as the position of the outline of the screen.
  • the screen position of the device can be simply and accurately located based on the boundary of the white background picture.
  • step S03 identifying the border of the white background picture from the photo, and using the border as the position of the outline of the screen, includes:
  • Step S031 converting the photo src into a grayscale image gray
  • Step S032 Specify a preset pixel threshold T1 to segment the grayscale image gray, wherein the pixel value of the pixel that exceeds the preset pixel threshold T1 in the photo src is set to 255, and the photo src The pixel value of the pixel in which the pixel does not exceed the preset pixel threshold T1 is set to 0;
  • Step S033 Obtain a continuous area of each pixel point in the gray image gray with a pixel value of 255;
  • a certain pixel is within 8 neighborhoods of another pixel, it can be considered that the two are continuous, and 2 or more continuous pixels can form a continuous area of pixels;
  • a pixel value of 0 is a black pixel, a pixel value of 255 represents a white pixel, and the connection area of a pixel with a pixel value of 0 does not need to be considered, and it is regarded as the background outside the screen area;
  • Step S034 Calculate the number of pixels in the continuous area of each pixel, and filter the continuous areas of each pixel, where the number of discarded pixels is less than the preset number threshold T2. Area, and reserve a continuous area where the number of pixels is greater than or equal to the preset number threshold T2;
  • step S036 the reserved pixel point continuous area whose fullness s is greater than the preset fullness threshold T3 is taken as the boundary of the white background picture, and the boundary is taken as the position of the outline of the screen.
  • the gray-scale image gray is segmented by specifying the preset pixel threshold T1; the number of pixels in the continuous area of each pixel is calculated, and the continuous area of each pixel is filtered; each reserved pixel is calculated The area of the minimum circumscribed rotating rectangle of the continuous area of pixels, calculate the fullness s of the minimum circumscribed rotated rectangle of each reserved pixel continuous area; the reserved pixels whose fullness s is greater than the preset fullness threshold T3 A continuous area is used as the boundary of the white background picture, and the boundary is used as the position of the outline of the screen, so as to accurately and reliably identify the screen positions of various terminals.
  • the present invention provides a screen scratch and chipping detection device, which includes:
  • Positioning device used to determine the contour position of the screen
  • the display shooting device is used to control the screen to display a full-screen yellow image below the preset exposure value, and shoot the yellow screen image based on the outline position of the screen; control the screen to display a full-screen black image higher than the preset exposure value, based on The outline position of the screen, taking a black screen image;
  • a feature extraction device for inputting the yellow screen image into a convolutional neural network, and extracting image features corresponding to the yellow screen image; inputting the black screen image into the convolutional neural network, and extracting the corresponding black screen image Image characteristics;
  • the convolutional neural network may be a resnext101 convolutional neural network to extract accurate image features
  • An identification device is used to obtain target candidate frames in the yellow screen image and the black screen image whose target categories are the scratch pattern category and the broken crack category based on the image features corresponding to the yellow screen image and the black screen image respectively.
  • the present invention obtains target candidate frames in the yellow screen image and the black screen image whose target categories are the scratch pattern category and the broken crack category based on the image features corresponding to the yellow screen image and the black screen image, respectively. It can accurately identify scratches or cracks on the screen of mobile phones and other devices, and can improve the efficiency of valuation and recycling of smart devices such as mobile phones.
  • the identification device is used for:
  • the corresponding yellow screen image corresponding to the multi-layer feature layer of different scales is obtained; based on the image features corresponding to the black screen image and Image features corresponding to the black screen image, and using the FPN method to obtain multi-layer feature layers of different scales corresponding to the black screen image;
  • the target candidate frame in the yellow screen image is extracted through the RPN network on the multi-layer feature layers of different scales corresponding to the yellow screen image, and each target candidate frame in the yellow screen image is preset to have scratches The probability value of grains and cracks;
  • the target candidate frame in the black screen image is extracted on the multi-layer feature layers of different scales corresponding to the black screen image through the RPN network, and the target candidate frame in the black screen image is preset The probability value of scratches and cracks in each target candidate frame;
  • the first 1000 target candidate frames in the yellow screen image with a larger probability value may be selected; the first 1000 target candidate frames in the black screen image with a larger probability value may be selected;
  • the first preset number of target candidate frames in the yellow screen image are input into the classification neural network, and each target candidate frame in the first preset number of target candidate frames in the yellow screen image corresponding to the output is obtained
  • the probability values belonging to the background category, the scratch pattern category and the broken crack category respectively input the previously preset number of target candidate frames in the black screen image into the classification neural network, and obtain the corresponding output black screen image
  • the classification neural network may be a fully connected layer classification neural network
  • the neural network outputs a target candidate frame a with a background category probability value of 0.2, a scratch pattern category has a probability value of 0.3, and a broken crack category has a probability value of 0.5, then the target candidate frame a
  • the initial category is the crack category
  • the neural network outputs the probability value of the background category of a certain target candidate frame b as 0.1, the probability value of the scratch pattern category is 0.2, and the probability value of the broken crack category is 0.7, then the initial value of the target candidate frame b The category is broken and cracked;
  • the initial category is determined as the target category of the target candidate frame
  • the preset probability threshold is 0.6
  • the neural network outputs that the initial category of a certain target candidate frame a is the broken crack category, and the probability value of the broken crack category is 0.5. Since the predetermined probability threshold of 0.6 is not exceeded, the initial category of the broken crack category of the target candidate frame a The category cannot be used as the target category;
  • the neural network outputs that the initial category of a certain target candidate frame b is the broken crack category, and the probability value of the broken crack category is 0.7. Since the preset probability threshold of 0.6 is not exceeded, the broken crack of the target candidate frame b The initial category of the category can be used as the target category;
  • the output determines the target candidate frame with the target category as scratch pattern category and broken crack category.
  • the initial category of the target candidate frame is determined, and then the target candidate frame of the determined target category is filtered from the target candidate frame of the determined initial category, which can further reliably and accurately identify the screen of the mobile phone and other devices. Scratches or cracks.
  • the identification device is used for:
  • the target candidate frames in the yellow screen image with the positions of the determined target categories overlapped are sorted in descending order based on the probability value to obtain the first sorting queue, and the target candidate frame with the highest probability value in the first sorting queue is used as the first reference Candidate frame, if the overlapping area of each target candidate frame in the subsequent queues in the first sorting queue and the first reference candidate frame exceeds the threshold of the area of the first reference candidate frame of the preset ratio, then The target candidate frame and its corresponding target category are deleted; the target candidate frames in the black screen image whose positions where the target category is determined overlap are arranged in descending order based on the probability value to obtain a second sorting queue, and the second sorting queue is placed in the second sorting queue.
  • the target candidate frame with the highest probability value is used as the second reference candidate frame. If the overlapping area of each subsequent target candidate frame in the second sorting queue and the second reference candidate frame exceeds the preset ratio of the second reference candidate frame The threshold of the area of, delete the target candidate frame and its corresponding target category;
  • the output determines the target candidate frame with the target category as scratch pattern category and broken crack category.
  • the preset ratio threshold may be 0.7.
  • the positioning device includes:
  • Display module used to display the bright screen of the screen as a white background picture
  • the screen may be a terminal device with a display screen, such as a mobile phone or a PAD;
  • the recognition module is configured to recognize the border of the white background picture from the photo, and use the border as the position of the outline of the screen.
  • the screen position of the device can be simply and accurately located based on the boundary of the white background picture.
  • the recognition module is configured to convert the photo src into a gray image gray; specify a preset pixel threshold T1 to segment the gray image gray, wherein, the pixel value of the pixel that exceeds the preset pixel threshold T1 in the photo src is set to 255, and the pixel value of the pixel that does not exceed the preset pixel threshold T1 in the photo src is set to 0 Obtain the continuous area of each pixel with a pixel value of 255 in the grayscale image gray; calculate the number of pixels in each continuous area of pixels, and screen each continuous area of pixels, where, Abandon the continuous area of pixels whose number of pixels is less than the preset number threshold T2, and reserve the continuous area of pixels where the number of pixels is greater than or equal to the preset number threshold T2; calculate that each reserved pixel is continuous The area of the minimum circumscribed rotating rectangle of the region of, calculate the fullness s of the minimum circumscribed rotated rectangle of each reserved
  • a certain pixel is within 8 neighborhoods of another pixel, it can be considered that the two are continuous, and 2 or more continuous pixels can form a continuous area of pixels;
  • a pixel value of 0 is a black pixel, a pixel value of 255 represents a white pixel, and the connection area of a pixel with a pixel value of 0 does not need to be considered, and it is regarded as the background outside the screen area;
  • the gray-scale image gray is segmented by specifying a preset pixel threshold T; the number of pixels in the continuous area of each pixel is calculated, and the continuous area of each pixel is filtered; each reservation is calculated The area of the minimum circumscribed rotating rectangle of the continuous area of pixels, calculate the fullness s of the minimum circumscribed rotated rectangle of each reserved pixel continuous area; the reserved pixels whose fullness s is greater than the preset fullness threshold T3 A continuous area is used as the boundary of the white background picture, and the boundary is used as the position of the outline of the screen, so as to accurately and reliably identify the screen positions of various terminals.
  • a computing-based device which includes:
  • a memory arranged to store computer-executable instructions which, when executed, cause the processor to:
  • Control the screen to display a full-screen yellow image below the preset exposure value, and take a yellow screen image based on the outline position of the screen;
  • Control the screen to display a full-screen black image higher than the preset exposure value, and take a black screen image based on the outline position of the screen;
  • target candidate frames in the yellow screen image and the black screen image are obtained in which the target categories are the scratch pattern category and the broken crack category.
  • a computer-readable storage medium having computer-executable instructions stored thereon, wherein, when the computer-executable instructions are executed by a processor, the processor:
  • Control the screen to display a full-screen yellow image below the preset exposure value, and take a yellow screen image based on the outline position of the screen;
  • Control the screen to display a full-screen black image higher than the preset exposure value, and take a black screen image based on the outline position of the screen;
  • target candidate frames in the yellow screen image and the black screen image are obtained in which the target categories are the scratch pattern category and the broken crack category.
  • the present invention can be implemented in software and/or a combination of software and hardware.
  • it can be implemented by an application specific integrated circuit (ASIC), a general purpose computer or any other similar hardware device.
  • the software program of the present invention may be executed by a processor to realize the above-mentioned steps or functions.
  • the software program (including related data structure) of the present invention can be stored in a computer-readable recording medium, such as a RAM memory, a magnetic or optical drive or a floppy disk and similar devices.
  • some steps or functions of the present invention may be implemented by hardware, for example, as a circuit that cooperates with a processor to execute each step or function.
  • a part of the present invention can be applied as a computer program product, such as a computer program instruction, when it is executed by a computer, through the operation of the computer, the method and/or technical solution according to the present invention can be invoked or provided.
  • the program instructions for invoking the method of the present invention may be stored in a fixed or removable recording medium, and/or transmitted through a data stream in a broadcast or other signal-bearing medium, and/or stored in accordance with the Said program instructions run in the working memory of the computer equipment.
  • an embodiment according to the present invention includes a device including a memory for storing computer program instructions and a processor for executing the program instructions, wherein, when the computer program instructions are executed by the processor, trigger
  • the operation of the device is based on the aforementioned methods and/or technical solutions according to multiple embodiments of the present invention.

Abstract

一种屏幕划痕碎裂检测方法及设备，所述方法通过分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框，可以准确识别出手机等设备屏幕上的划痕纹或碎裂纹，可以提高手机等智能设备估价回收等效率。

Description

屏幕划痕碎裂检测方法及设备 技术领域
本发明涉及计算机领域，尤其涉及一种屏幕划痕碎裂检测方法及设备。
背景技术
现有的手机等设备等屏幕划痕碎裂检测方式都是人工方式，费时费力，影响手机等智能设备估价回收等效率。
发明内容
本发明的一个目的是提供一种屏幕划痕碎裂检测方法及设备。
根据本发明的一个方面，提供了一种屏幕划痕碎裂检测方法，该方法包括：
确定屏幕的轮廓位置；
控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
进一步的，上述方法中，所述卷积神经网络为resnext101卷积神经 网络。
进一步的，上述方法中，分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中，目标类别为划痕纹类别和碎裂纹类别的目标候选框，包括：
基于所述黄色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黄色屏幕图像对应的不同尺度的多层特征层；基于所述黑色屏幕图像对应的图像特征和黑色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黑色屏幕图像对应的不同尺度的多层特征层；
通过RPN网络在所述黄色屏幕图像对应的不同尺度的多层特征层进行所述黄色屏幕图像中的目标候选框的提取，并预设所述黄色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；通过RPN网络在所述黑色屏幕图像对应的不同尺度的多层特征层进行所述黑色屏幕图像中的目标候选框的提取，并预设所述黑色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；
选取概率值较大的所述黄色屏幕图像中的前预设个数的目标候选框；选取概率值较大的所述黑色屏幕图像中的前预设个数的目标候选框；
将所述黄色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黄色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；将所述黑色色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黑色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；
将每个目标候选框的概率值较大的对应类别确定为该目标候选框的初始类别；
若确定初始类别的目标候选框的该初始类别的概率值大于预设概率阈值，则将该初始类别确定为该目标候选框的目标类别；
输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
进一步的，上述方法中，输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框，包括：
对所述黄色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第一排序队列，将所述第一排序队列中概率值最高的目标候选框作为第一基准候选框，若所述第一排序队列中的后续队列中的每一个目标候选框与所述第一基准候选框的重叠面积是超过预设比例的第一基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；对所述黑色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第二排序队列，将所述第二排序队列中概率值最高的目标候选框作为第二基准候选框，若所述第二排序队列中的后续每一个目标候选框与所述第二基准候选框的重叠面积超过预设比例的第二基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；
输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
进一步的，上述方法中，所述分类神经网络为全连接层分类神经网络。
进一步的，上述方法中，确定屏幕的轮廓位置，包括：
将屏幕亮屏显示为白底画面；
拍摄包括所述白底画面的屏幕的照片；
从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
进一步的，上述方法中，从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置，包括：
将所述照片转换为灰度图片；
指定预设像素阈值T1对所述灰度图片进行分割，其中，将所述照片中超过所述预设像素阈值T1的像素点的像素值设为255，将所述照片中未 超过所述预设像素阈值T1的像素点的像素值设为0；
获取所述灰度图片中的像素值为255的各个像素点连续的区域；
计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选，其中，舍弃像素点的个数量小于预设个数阈值T2的像素点连续的区域，并保留像素点的个数量大于等于预设个数阈值T2的像素点连续的区域；
计算每个保留的像素点连续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s，其中，饱满度s＝某个保留的像素点连续的区域中的像素点的个数/该个保留的像素点连续的区域的最小外接旋转矩形的面积；
将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
根据本发明的另一方面，还提供一种屏幕划痕碎裂检测设备，其中，该设备包括：
定位装置，用于确定屏幕的轮廓位置；
显示拍摄装置，用于控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
特征提取装置，用于将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
识别装置，用于分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
进一步的，上述设备中，所述卷积神经网络为resnext101卷积神经 网络。
进一步的，上述设备中，所述识别装置，用于：
基于所述黄色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黄色屏幕图像对应的不同尺度的多层特征层；基于所述黑色屏幕图像对应的图像特征和黑色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黑色屏幕图像对应的不同尺度的多层特征层；
通过RPN网络在所述黄色屏幕图像对应的不同尺度的多层特征层进行所述黄色屏幕图像中的目标候选框的提取，并预设所述黄色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；通过RPN网络在所述黑色屏幕图像对应的不同尺度的多层特征层进行所述黑色屏幕图像中的目标候选框的提取，并预设所述黑色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；
选取概率值较大的所述黄色屏幕图像中的前预设个数的目标候选框；选取概率值较大的所述黑色屏幕图像中的前预设个数的目标候选框；
将所述黄色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黄色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；将所述黑色色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黑色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；
将每个目标候选框的概率值较大的对应类别确定为该目标候选框的初始类别；
若确定初始类别的目标候选框的该初始类别的概率值大于预设概率阈值，则将该初始类别确定为该目标候选框的目标类别；
输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
进一步的，上述设备中，所述识别装置，用于：
对所述黄色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第一排序队列，将所述第一排序队列中概率值最高的目标候选框作为第一基准候选框，若所述第一排序队列中的后续队列中的每一个目标候选框与所述第一基准候选框的重叠面积是超过预设比例的第一基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；对所述黑色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第二排序队列，将所述第二排序队列中概率值最高的目标候选框作为第二基准候选框，若所述第二排序队列中的后续每一个目标候选框与所述第二基准候选框的重叠面积超过预设比例的第二基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；
输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
进一步的，上述设备中，所述分类神经网络为全连接层分类神经网络。
进一步的，上述设备中，所述定位装置，包括：
显示模块，用于将屏幕亮屏显示为白底画面；
拍摄模块，用于拍摄包括所述白底画面的屏幕的照片；
识别模块，用于从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
进一步的，上述设备中，所述识别模块，用于将所述照片转换为灰度图片；指定预设像素阈值T1对所述灰度图片进行分割，其中，将所述照片中超过所述预设像素阈值T1的像素点的像素值设为255，将所述照片中未超过所述预设像素阈值T1的像素点的像素值设为0；获取所述灰度图片中像素值为255的各个像素点连续的区域；计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选，其中，舍弃像素点的个数量小于预设个数阈值T2的像素点连续的区域，并保留像素点的个数量大于等于预设个数阈值T2的像素点连续的区域；计算每个保留的像素点连 续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s，其中，饱满度s＝某个保留的像素点连续的区域中的像素点的个数/该个保留的像素点连续的区域的最小外接旋转矩形的面积；将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
根据本发明的另一方面，还提供一种基于计算的设备，其中，包括：
处理器；以及
被安排成存储计算机可执行指令的存储器，所述可执行指令在被执行时使所述处理器：
确定屏幕的轮廓位置；
控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
根据本发明的另一方面，还提供一种计算机可读存储介质，其上存储有计算机可执行指令，其中，该计算机可执行指令被处理器执行时使得该处理器：
确定屏幕的轮廓位置；
控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
与现有技术相比，本发明通过分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框，可以准确识别出手机等设备屏幕上的划痕纹或碎裂纹，可以提高手机等智能设备估价回收等效率。
附图说明
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述，本发明的其它特征、目的和优点将会变得更明显：
图1示出本发明一实施例的屏幕划痕碎裂检测方法的流程图。
附图中相同或相似的附图标记代表相同或相似的部件。
具体实施方式
下面结合附图对本发明作进一步详细描述。
在本申请一个典型的配置中，终端、服务网络的设备和可信方均包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器，随机存取存储器 (RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括非暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
如图1所示，本发明提供一种屏幕划痕碎裂检测方法，所述方法包括：
步骤S0，确定屏幕的轮廓位置；
步骤S1,控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
在此，针对白色轮廓的设备如手机、PAD等，通过获取黄色屏幕图像，可以保证此类设备的屏幕上的划痕纹、碎裂纹的识别准确度；
步骤S2,控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
在此，拍摄高、低曝光值图片的目的是：高曝光值图片有利于拍摄深色屏幕表面纹路，但对于亮色屏幕表面纹路容易产生过曝问题，因此需要使用低曝光值图片辅助检测；
拍摄黑、黄两种图片的目的是：经过实验得知不同类型的纹路在不同背景颜色图片拍摄下清晰程度不同，因此我们选用了实验效果较好的黑色 和黄色图片作为背景；
步骤S3,将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
在此，所述卷积神经网络可以是resnext101卷积神经网络，以提取到准确的图像特征；
步骤S4，分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
在此，本发明通过分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框，可以准确识别出手机等设备屏幕上的划痕纹或碎裂纹，可以提高手机等智能设备估价回收等效率。
本发明的屏幕划痕碎裂检测方法一实施例中，步骤S4，分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中，目标类别为划痕纹类别和碎裂纹类别的目标候选框，包括：
步骤S41，基于所述黄色屏幕图像对应的图像特征，并通过FPN(feature pyramid networks)方法，得到对应的所述黄色屏幕图像对应的不同尺度的多层特征层；基于所述黑色屏幕图像对应的图像特征和黑色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黑色屏幕图像对应的不同尺度的多层特征层；
步骤S42，通过RPN(Region Proposal Network)网络在所述黄色屏幕图像对应的不同尺度的多层特征层进行所述黄色屏幕图像中的目标候选框的提取，并预设所述黄色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；通过RPN网络在所述黑色屏幕图像对应的不同尺度的多层特 征层进行所述黑色屏幕图像中的目标候选框的提取，并预设所述黑色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；
步骤S43，选取概率值较大的所述黄色屏幕图像中的前预设个数的目标候选框；选取概率值较大的所述黑色屏幕图像中的前预设个数的目标候选框；
在此，可以选取概率值较大的所述黄色屏幕图像中的前1000个的目标候选框；选取概率值较大的所述黑色屏幕图像中的前1000个的目标候选框；
步骤S44，将所述黄色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黄色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；将所述黑色色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黑色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；
在此，所述分类神经网络可以是全连接层分类神经网络，以得到可靠等分类；
步骤S45，将每个目标候选框的概率值较大的对应类别确定为该目标候选框的初始类别；
在此，例如，所述神经网络输出某个目标候选框a的背景类别的概率值为0.2，划痕纹类别的概率值为0.3，碎裂纹类别的概率值为0.5，那么该目标候选框a的初始类别为碎裂纹类别；
又如，所述神经网络输出某个目标候选框b的背景类别的概率值为0.1，划痕纹类别的概率值为0.2，碎裂纹类别的概率值为0.7，那么该目标候选框b的初始类别为碎裂纹类别；
步骤S46，若确定初始类别的目标候选框的该初始类别的概率值大于预设概率阈值，则将该初始类别确定为该目标候选框的目标类别；
在此，例如，预设概率阈值为0.6，
所述神经网络输出某个目标候选框a的初始类别为碎裂纹类别，碎裂纹类别的概率值为0.5，由于没有超过0.6的预设概率阈值，所以该目标候选框a的碎裂纹类别的初始类别不能作为目标类别；
又如，所述神经网络输出某个目标候选框b的初始类别为碎裂纹类别，碎裂纹类别的概率值为0.7，由于超过0.6的预设概率阈值，所以该目标候选框b的碎裂纹类别的初始类别可以作为目标类别；
步骤S47，输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
在此，本实施例通过目标候选框的初始类别的确定，再从确定初始类别的目标候选框中筛选出确定目标类别的目标候选框，能够进一步可靠、准确的识别出手机等设备屏幕上的划痕纹或碎裂纹。
本发明的屏幕划痕碎裂检测方法一实施例中，步骤S47，输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框，包括：
步骤S471，对所述黄色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第一排序队列，将所述第一排序队列中概率值最高的目标候选框作为第一基准候选框，若所述第一排序队列中的后续队列中的每一个目标候选框与所述第一基准候选框的重叠面积是超过预设比例的第一基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；对所述黑色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第二排序队列，将所述第二排序队列中概率值最高的目标候选框作为第二基准候选框，若所述第二排序队列中的后续每一个目标候选框与所述第二基准候选框的重叠面积超过预设比例的第二基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；
步骤S472，输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选 框。
在此，所述预设比例阈值可以为0.7，当所述排序队列中的后续队列中的每一个目标候选框与所述基准候选框的重叠面积超过0.7的比例所述基准候选框的面积，则将目标候选框及其对应的目标类别删除；
本实施例通过将重叠面积超过预设比例的基准候选框的面积准候选框的面积的阈值的后续每一个目标候选框进行进一步过滤删除，可以保证输出的可靠的确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
本发明的屏幕划痕碎裂检测方法一实施例中，步骤S0，确定屏幕的轮廓位置，包括：
步骤S01，将屏幕亮屏显示为白底画面；
在此，所述屏幕可以是手机、PAD等带有显示屏幕的终端设备；
步骤S02，拍摄包括所述白底画面的屏幕的照片；
在此，拍摄屏幕的时候，会同时把屏幕区域之外的无关区域也拍摄进去，后续需要从中识别屏幕区域；
步骤S03，从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
在此，本发明通过将屏幕亮屏显示为白底画面，基于白底画面的边界可以简单、准确的定位设备的屏幕位置。
本发明的屏幕定位方法一实施例中，步骤S03，从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置，包括：
步骤S031，将所述照片src转换为灰度图片gray；
步骤S032，指定预设像素阈值T1对所述灰度图片gray进行分割，其中，将所述照片src中超过所述预设像素阈值T1的像素点的像素值设为255，将所述照片src中未超过所述预设像素阈值T1的像素点的像素值值设为0；
步骤S033，获取所述灰度图片gray中的像素值为255的各个像素点 连续的区域；
在此，某个像素点在另一个像素点的8邻域内，可以认为两者是连续的，2个或2个以上连续像素点可以形成一个像素点连续的区域；
像素值为0为黑色的像素点，像素值为255表示白色的像素点，像素值为0的像素点的连接区域不需要考虑，视为屏幕区域之外的背景；
步骤S034，计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选，其中，舍弃像素点的个数量小于预设个数阈值T2的像素点连续的区域，并保留像素点的个数量大于等于预设个数阈值T2的像素点连续的区域；
步骤S035，计算每个保留的像素点连续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s，其中，饱满度s＝某个保留的像素点连续的区域中的像素点的个数/该个保留的像素点连续的区域的最小外接旋转矩形的面积；
步骤S036，将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
在此，可以遍历每个保留的像素点连续的区域，用每个保留的像素点连续的区域的像素点个数除以其最小外接旋转矩形的面积，得到该区域的饱满度s，如果某个保留的像素点连续的区域的饱满度s值大于预设饱满度阈值T3则其为屏幕区域，若小于，则为非屏幕区域。
本实施通过指定预设像素阈值T1对所述灰度图片gray进行分割；计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选；计算每个保留的像素点连续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s；将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置，从而准确、可靠的识别出各种终端的屏幕位置。
本发明提供一种屏幕划痕碎裂检测设备，所述设备包括：
定位装置，用于确定屏幕的轮廓位置；
显示拍摄装置,用于控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
在此，针对白色轮廓的设备如手机、PAD等，通过获取黄色屏幕图像，可以保证此类设备的屏幕上的划痕纹、碎裂纹的识别准确度；
特征提取装置,用于将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
在此，所述卷积神经网络可以是resnext101卷积神经网络，以提取到准确的图像特征；
识别装置，用于分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
在此，本发明通过分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框，可以准确识别出手机等设备屏幕上的划痕纹或碎裂纹，可以提高手机等智能设备估价回收等效率。
本发明的屏幕划痕碎裂检测设备一实施例中，所述识别装置，用于：
基于所述黄色屏幕图像对应的图像特征，并通过FPN(feature pyramid networks)方法，得到对应的所述黄色屏幕图像对应的不同尺度的多层特征层；基于所述黑色屏幕图像对应的图像特征和黑色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黑色屏幕图像对应的不同尺度的多层特征层；
通过RPN网络在所述黄色屏幕图像对应的不同尺度的多层特征层进行所述黄色屏幕图像中的目标候选框的提取，并预设所述黄色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；通过RPN网络在所述黑色屏幕图像对应的不同尺度的多层特征层进行所述黑色屏幕图像中的目标候选框的提取，并预设所述黑色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；
选取概率值较大的所述黄色屏幕图像中的前预设个数的目标候选框；选取概率值较大的所述黑色屏幕图像中的前预设个数的目标候选框；
在此，可以选取概率值较大的所述黄色屏幕图像中的前1000个的目标候选框；选取概率值较大的所述黑色屏幕图像中的前1000个的目标候选框；
将所述黄色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黄色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；将所述黑色色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黑色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；
在此，所述分类神经网络可以是全连接层分类神经网络；
将每个目标候选框的概率值较大的对应类别确定为该目标候选框的初始类别；
在此，例如，所述神经网络输出某个目标候选框a的背景类别的概率值为0.2，划痕纹类别的概率值为0.3，碎裂纹类别的概率值为0.5，那么该目标候选框a的初始类别为碎裂纹类别；
又如，所述神经网络输出某个目标候选框b的背景类别的概率值为0.1，划痕纹类别的概率值为0.2，碎裂纹类别的概率值为0.7，那么该目标候选框b的初始类别为碎裂纹类别；
若确定初始类别的目标候选框的该初始类别的概率值大于预设概率阈 值，则将该初始类别确定为该目标候选框的目标类别；
在此，例如，预设概率阈值为0.6，
所述神经网络输出某个目标候选框a的初始类别为碎裂纹类别，碎裂纹类别的概率值为0.5，由于没有超过0.6的预设概率阈值，所以该目标候选框a的碎裂纹类别的初始类别不能作为目标类别；
又如，所述神经网络输出某个目标候选框b的初始类别为碎裂纹类别，碎裂纹类别的概率值为0.7，由于没有超过0.6的预设概率阈值，所以该目标候选框b的碎裂纹类别的初始类别可以作为目标类别；
输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
在此，本实施例通过目标候选框的初始类别的确定，再从确定初始类别的目标候选框中筛选出确定目标类别的目标候选框，能够进一步可靠、准确的识别出手机等设备屏幕上的划痕纹或碎裂纹。
本发明的屏幕划痕碎裂检测设备一实施例中，所述识别装置，用于：
对所述黄色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第一排序队列，将所述第一排序队列中概率值最高的目标候选框作为第一基准候选框，若所述第一排序队列中的后续队列中的每一个目标候选框与所述第一基准候选框的重叠面积是超过预设比例的第一基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；对所述黑色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第二排序队列，将所述第二排序队列中概率值最高的目标候选框作为第二基准候选框，若所述第二排序队列中的后续每一个目标候选框与所述第二基准候选框的重叠面积超过预设比例的第二基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；
输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
在此，所述预设比例阈值可以为0.7，当所述排序队列中的后续队列 中的每一个目标候选框与所述基准候选框的重叠面积超过0.7的比例所述基准候选框的面积，则将目标候选框及其对应的目标类别删除；
本实施例通过将重叠面积超过预设比例的基准候选框的面积准候选框的面积的阈值的后续每一个目标候选框进行进一步过滤删除，可以保证输出的可靠的确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
本发明的屏幕划痕碎裂检测设备一实施例中，所述定位装置，包括：
显示模块，用于将屏幕亮屏显示为白底画面；
在此，所述屏幕可以是手机、PAD等带有显示屏幕的终端设备；
拍摄模块，用于拍摄包括所述白底画面的屏幕的照片；
在此，拍摄屏幕的时候，会同时把屏幕区域之外的无关区域也拍摄进去，后续需要从中识别屏幕区域；
识别模块，用于从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
在此，本发明通过将屏幕亮屏显示为白底画面，基于白底画面的边界可以简单、准确的定位设备的屏幕位置。
本发明的屏幕划痕碎裂检测设备一实施例中，所述识别模块，用于将所述照片src转换为灰度图片gray；指定预设像素阈值T1对所述灰度图片gray进行分割，其中，将所述照片src中超过所述预设像素阈值T1的像素点的像素值设为255，将所述照片src中未超过所述预设像素阈值T1的像素点的像素值设为0；获取所述灰度图片gray中像素值为255的各个像素点连续的区域；计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选，其中，舍弃像素点的个数量小于预设个数阈值T2的像素点连续的区域，并保留像素点的个数量大于等于预设个数阈值T2的像素点连续的区域；计算每个保留的像素点连续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s，其中，饱满度s＝某个保留的像素点连续的区域中的像素点的个 数/该个保留的像素点连续的区域的最小外接旋转矩形的面积；将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
在此，某个像素点在另一个像素点的8邻域内，可以认为两者是连续的，2个或2个以上连续像素点可以形成一个像素点连续的区域；
像素值为0为黑色的像素点，像素值为255表示白色的像素点，像素值为0的像素点的连接区域不需要考虑，视为屏幕区域之外的背景；
可以遍历每个保留的像素点连续的区域，用每个保留的像素点连续的区域的像素点个数除以其最小外接旋转矩形的面积，得到该区域的饱满度s，如果某个保留的像素点连续的区域的饱满度s值大于预设饱满度阈值T3则其为屏幕区域，若小于，则为非屏幕区域。
本实施通过指定预设像素阈值T对所述灰度图片gray进行分割；计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选；计算每个保留的像素点连续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s；将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置，从而准确、可靠的识别出各种终端的屏幕位置。
根据本发明的另一方面，还提供一种基于计算的设备，其中，包括：
处理器；以及
被安排成存储计算机可执行指令的存储器，所述可执行指令在被执行时使所述处理器：
确定屏幕的轮廓位置；
控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
根据本发明的另一方面，还提供一种计算机可读存储介质，其上存储有计算机可执行指令，其中，该计算机可执行指令被处理器执行时使得该处理器：
确定屏幕的轮廓位置；
控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
本发明的各设备和存储介质实施例的详细内容，具体可参见各方法实施例的对应部分，在此，不再赘述。
显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离 本申请的精神和范围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。
需要注意的是，本发明可在软件和/或软件与硬件的组合体中被实施，例如，可采用专用集成电路(ASIC)、通用目的计算机或任何其他类似硬件设备来实现。在一个实施例中，本发明的软件程序可以通过处理器执行以实现上文所述步骤或功能。同样地，本发明的软件程序(包括相关的数据结构)可以被存储到计算机可读记录介质中，例如，RAM存储器，磁或光驱动器或软磁盘及类似设备。另外，本发明的一些步骤或功能可采用硬件来实现，例如，作为与处理器配合从而执行各个步骤或功能的电路。
另外，本发明的一部分可被应用为计算机程序产品，例如计算机程序指令，当其被计算机执行时，通过该计算机的操作，可以调用或提供根据本发明的方法和/或技术方案。而调用本发明的方法的程序指令，可能被存储在固定的或可移动的记录介质中，和/或通过广播或其他信号承载媒体中的数据流而被传输，和/或被存储在根据所述程序指令运行的计算机设备的工作存储器中。在此，根据本发明的一个实施例包括一个装置，该装置包括用于存储计算机程序指令的存储器和用于执行程序指令的处理器，其中，当该计算机程序指令被该处理器执行时，触发该装置运行基于前述根据本发明的多个实施例的方法和/或技术方案。
对于本领域技术人员而言，显然本发明不限于上述示范性实施例的细节，而且在不背离本发明的精神或基本特征的情况下，能够以其他的具体形式实现本发明。因此，无论从哪一点来看，均应将实施例看作是示范性的，而且是非限制性的，本发明的范围由所附权利要求而不是上述说明限定，因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本发明内。不应将权利要求中的任何附图标记视为限制所涉及的权利要求。此外，显然“包括”一词不排除其他单元或步骤，单数不排除复数。 装置权利要求中陈述的多个单元或装置也可以由一个单元或装置通过软件或者硬件来实现。第一，第二等词语用来表示名称，而并不表示任何特定的顺序。

Claims (16)

  1. 一种屏幕划痕碎裂检测方法，其中，该方法包括：
    确定屏幕的轮廓位置；
    控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
    控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
    将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
    分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
  2. 根据权利要求1所述的方法，其中，所述卷积神经网络为resnext101卷积神经网络。
  3. 根据权利要求1所述的方法，其中，分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中，目标类别为划痕纹类别和碎裂纹类别的目标候选框，包括：
    基于所述黄色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黄色屏幕图像对应的不同尺度的多层特征层；基于所述黑色屏幕图像对应的图像特征和黑色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黑色屏幕图像对应的不同尺度的多层特征层；
    通过RPN网络在所述黄色屏幕图像对应的不同尺度的多层特征层进行 所述黄色屏幕图像中的目标候选框的提取，并预设所述黄色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；通过RPN网络在所述黑色屏幕图像对应的不同尺度的多层特征层进行所述黑色屏幕图像中的目标候选框的提取，并预设所述黑色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；
    选取概率值较大的所述黄色屏幕图像中的前预设个数的目标候选框；选取概率值较大的所述黑色屏幕图像中的前预设个数的目标候选框；
    将所述黄色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黄色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；将所述黑色色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黑色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；
    将每个目标候选框的概率值较大的对应类别确定为该目标候选框的初始类别；
    若确定初始类别的目标候选框的该初始类别的概率值大于预设概率阈值，则将该初始类别确定为该目标候选框的目标类别；
    输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
  4. 根据权利要求3所述的方法，其中，输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框，包括：
    对所述黄色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第一排序队列，将所述第一排序队列中概率值最高的目标候选框作为第一基准候选框，若所述第一排序队列中的后续队 列中的每一个目标候选框与所述第一基准候选框的重叠面积是超过预设比例的第一基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；对所述黑色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第二排序队列，将所述第二排序队列中概率值最高的目标候选框作为第二基准候选框，若所述第二排序队列中的后续每一个目标候选框与所述第二基准候选框的重叠面积超过预设比例的第二基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；
    输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
  5. 根据权利要求1所述的方法，其中，所述分类神经网络为全连接层分类神经网络。
  6. 根据权利要求1所述的方法，其中，确定屏幕的轮廓位置，包括：
    将屏幕亮屏显示为白底画面；
    拍摄包括所述白底画面的屏幕的照片；
    从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
  7. 根据权利要求6所述的方法，其中，从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置，包括：
    将所述照片转换为灰度图片；
    指定预设像素阈值T1对所述灰度图片进行分割，其中，将所述照片中超过所述预设像素阈值T1的像素点的像素值设为255，将所述照片中未超过所述预设像素阈值T1的像素点的像素值设为0；
    获取所述灰度图片中的像素值为255的各个像素点连续的区域；
    计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选，其中，舍弃像素点的个数量小于预设个数阈值T2的像素点连续的区域，并保留像素点的个数量大于等于预设个数阈值T2的像素点连续的区域；
    计算每个保留的像素点连续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s，其中，饱满度s＝某个保留的像素点连续的区域中的像素点的个数/该个保留的像素点连续的区域的最小外接旋转矩形的面积；
    将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
  8. 一种屏幕划痕碎裂检测设备，其中，该设备包括：
    定位装置，用于确定屏幕的轮廓位置；
    显示拍摄装置，用于控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
    特征提取装置，用于将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
    识别装置，用于分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
  9. 根据权利要求8所述的设备，其中，所述卷积神经网络为 resnext101卷积神经网络。
  10. 根据权利要求8所述的设备，其中，所述识别装置，用于：
    基于所述黄色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黄色屏幕图像对应的不同尺度的多层特征层；基于所述黑色屏幕图像对应的图像特征和黑色屏幕图像对应的图像特征，并通过FPN方法，得到对应的所述黑色屏幕图像对应的不同尺度的多层特征层；
    通过RPN网络在所述黄色屏幕图像对应的不同尺度的多层特征层进行所述黄色屏幕图像中的目标候选框的提取，并预设所述黄色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；通过RPN网络在所述黑色屏幕图像对应的不同尺度的多层特征层进行所述黑色屏幕图像中的目标候选框的提取，并预设所述黑色屏幕图像中的每个目标候选框存在划痕纹和碎裂纹的概率值；
    选取概率值较大的所述黄色屏幕图像中的前预设个数的目标候选框；选取概率值较大的所述黑色屏幕图像中的前预设个数的目标候选框；
    将所述黄色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黄色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；将所述黑色色屏幕图像中的前预设个数的目标候选框输入分类神经网络，并获取对应输出的所述黑色屏幕图像中的前预设个数的目标候选框中的每一目标候选框分别属于背景类别、划痕纹类别和碎裂纹类别的概率值；
    将每个目标候选框的概率值较大的对应类别确定为该目标候选框的初始类别；
    若确定初始类别的目标候选框的该初始类别的概率值大于预设概率阈值，则将该初始类别确定为该目标候选框的目标类别；
    输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
  11. 根据权利要求10所述的设备，其中，所述识别装置，用于：
    对所述黄色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第一排序队列，将所述第一排序队列中概率值最高的目标候选框作为第一基准候选框，若所述第一排序队列中的后续队列中的每一个目标候选框与所述第一基准候选框的重叠面积是超过预设比例的第一基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；对所述黑色屏幕图像中的确定了目标类别的位置重叠的目标候选框基于概率值进行降序排列得到第二排序队列，将所述第二排序队列中概率值最高的目标候选框作为第二基准候选框，若所述第二排序队列中的后续每一个目标候选框与所述第二基准候选框的重叠面积超过预设比例的第二基准候选框的面积的阈值，则将目标候选框及其对应的目标类别删除；
    输出确定了目标类别为划痕纹类别和碎裂纹类别的目标候选框。
  12. 根据权利要求10所述的设备，其中，所述分类神经网络为全连接层分类神经网络。
  13. 根据权利要求8所述的设备，其中，所述定位装置，包括：
    显示模块，用于将屏幕亮屏显示为白底画面；
    拍摄模块，用于拍摄包括所述白底画面的屏幕的照片；
    识别模块，用于从所述照片中识别出所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
  14. 根据权利要求13所述的装置，其中，所述识别模块，用于将所 述照片转换为灰度图片；指定预设像素阈值T1对所述灰度图片进行分割，其中，将所述照片中超过所述预设像素阈值T1的像素点的像素值设为255，将所述照片中未超过所述预设像素阈值T1的像素点的像素值设为0；获取所述灰度图片中像素值为255的各个像素点连续的区域；计算每个像素点连续的区域中的像素点的个数，对每个像素点连续的区域进行筛选，其中，舍弃像素点的个数量小于预设个数阈值T2的像素点连续的区域，并保留像素点的个数量大于等于预设个数阈值T2的像素点连续的区域；计算每个保留的像素点连续的区域的最小外接旋转矩形的面积，计算每个保留的像素点连续的区域的最小外接旋转矩形的饱满度s，其中，饱满度s＝某个保留的像素点连续的区域中的像素点的个数/该个保留的像素点连续的区域的最小外接旋转矩形的面积；将饱满度s大于预设饱满度阈值T3的保留的像素点连续的区域作为所述白底画面的边界，将所述边界作为所述屏幕的轮廓的位置。
  15. 一种基于计算的设备，其中，包括：
    处理器；以及
    被安排成存储计算机可执行指令的存储器，所述可执行指令在被执行时使所述处理器：
    确定屏幕的轮廓位置；
    控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
    控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
    将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
    分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
  16. 一种计算机可读存储介质，其上存储有计算机可执行指令，其中，该计算机可执行指令被处理器执行时使得该处理器：
    确定屏幕的轮廓位置；
    控制屏幕显示低于预设曝光值的满屏黄色的图像，基于屏幕的轮廓位置，拍摄黄色屏幕图像；
    控制屏幕显示高于预设曝光值的满屏黑色的图像，基于屏幕的轮廓位置，拍摄黑色屏幕图像；
    将所述黄色屏幕图像输入卷积神经网络，提取到所述黄色屏幕图像对应的图像特征；将所述黑色屏幕图像输入卷积神经网络，提取到所述黑色屏幕图像对应的图像特征；
    分别基于所述黄色屏幕图像和黑色屏幕图像对应的图像特征，得到所述黄色屏幕图像和黑色屏幕图像中的目标类别为划痕纹类别和碎裂纹类别的目标候选框。
PCT/CN2020/120892 2020-01-21 2020-10-14 屏幕划痕碎裂检测方法及设备 WO2021147387A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010072749.2 2020-01-21
CN202010072749.2A CN111325716B (zh) 2020-01-21 2020-01-21 屏幕划痕碎裂检测方法及设备

Publications (1)

Publication Number Publication Date
WO2021147387A1 true WO2021147387A1 (zh) 2021-07-29

Family

ID=71172526

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/120892 WO2021147387A1 (zh) 2020-01-21 2020-10-14 屏幕划痕碎裂检测方法及设备

Country Status (2)

Country Link
CN (1) CN111325716B (zh)
WO (1) WO2021147387A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114663418A (zh) * 2022-04-06 2022-06-24 京东安联财产保险有限公司 图像处理的方法及装置、存储介质及电子设备
US11436570B2 (en) 2014-10-31 2022-09-06 Ecoatm, Llc Systems and methods for recycling consumer electronic devices
US11462868B2 (en) 2019-02-12 2022-10-04 Ecoatm, Llc Connector carrier for electronic device kiosk
US11482067B2 (en) 2019-02-12 2022-10-25 Ecoatm, Llc Kiosk for evaluating and purchasing used electronic devices
CN115511907A (zh) * 2022-11-24 2022-12-23 深圳市晶台股份有限公司 一种用于led屏幕的划痕检测方法
CN116452613A (zh) * 2023-06-14 2023-07-18 山东省国土空间生态修复中心(山东省地质灾害防治技术指导中心、山东省土地储备中心) 一种地质调查中裂缝轮廓提取方法
US11798250B2 (en) 2019-02-18 2023-10-24 Ecoatm, Llc Neural network based physical condition evaluation of electronic devices, and associated systems and methods
US11922467B2 (en) 2020-08-17 2024-03-05 ecoATM, Inc. Evaluating an electronic device using optical character recognition

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111325716B (zh) * 2020-01-21 2023-09-01 上海万物新生环保科技集团有限公司 屏幕划痕碎裂检测方法及设备
CN111175318A (zh) * 2020-01-21 2020-05-19 上海悦易网络信息技术有限公司 屏幕划痕碎裂检测方法及设备
CN111885262B (zh) * 2020-07-24 2021-08-03 广州绿怡信息科技有限公司 一种基于智能终端机上的手机投保方法
CN113109368B (zh) * 2021-03-12 2023-09-01 浙江华睿科技股份有限公司 玻璃裂纹检测方法、装置、设备及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108593672A (zh) * 2018-03-01 2018-09-28 深圳回收宝科技有限公司 一种终端触控屏的检测方法、检测装置以及存储介质
US20190004543A1 (en) * 2017-07-03 2019-01-03 Skydio, Inc. Detecting optical discrepancies in captured images
CN109919002A (zh) * 2019-01-23 2019-06-21 平安科技（深圳）有限公司 黄色禁停线识别方法、装置、计算机设备及存储介质
CN111175318A (zh) * 2020-01-21 2020-05-19 上海悦易网络信息技术有限公司 屏幕划痕碎裂检测方法及设备
CN111325716A (zh) * 2020-01-21 2020-06-23 上海悦易网络信息技术有限公司 屏幕划痕碎裂检测方法及设备

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108280822B (zh) * 2017-12-20 2020-08-25 歌尔科技有限公司 屏幕划痕的检测方法及装置
CN109765245B (zh) * 2019-02-25 2021-08-13 武汉精立电子技术有限公司 大尺寸显示屏缺陷检测定位方法
CN110084801A (zh) * 2019-04-28 2019-08-02 深圳回收宝科技有限公司 一种终端屏幕的检测方法、装置、便携式终端和存储介质
CN110222787B (zh) * 2019-06-14 2022-10-14 合肥工业大学 多尺度目标检测方法、装置、计算机设备及存储介质
CN110675399A (zh) * 2019-10-28 2020-01-10 上海悦易网络信息技术有限公司 屏幕外观瑕疵检测方法及设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190004543A1 (en) * 2017-07-03 2019-01-03 Skydio, Inc. Detecting optical discrepancies in captured images
CN108593672A (zh) * 2018-03-01 2018-09-28 深圳回收宝科技有限公司 一种终端触控屏的检测方法、检测装置以及存储介质
CN109919002A (zh) * 2019-01-23 2019-06-21 平安科技（深圳）有限公司 黄色禁停线识别方法、装置、计算机设备及存储介质
CN111175318A (zh) * 2020-01-21 2020-05-19 上海悦易网络信息技术有限公司 屏幕划痕碎裂检测方法及设备
CN111325716A (zh) * 2020-01-21 2020-06-23 上海悦易网络信息技术有限公司 屏幕划痕碎裂检测方法及设备

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11436570B2 (en) 2014-10-31 2022-09-06 Ecoatm, Llc Systems and methods for recycling consumer electronic devices
US11462868B2 (en) 2019-02-12 2022-10-04 Ecoatm, Llc Connector carrier for electronic device kiosk
US11482067B2 (en) 2019-02-12 2022-10-25 Ecoatm, Llc Kiosk for evaluating and purchasing used electronic devices
US11843206B2 (en) 2019-02-12 2023-12-12 Ecoatm, Llc Connector carrier for electronic device kiosk
US11798250B2 (en) 2019-02-18 2023-10-24 Ecoatm, Llc Neural network based physical condition evaluation of electronic devices, and associated systems and methods
US11922467B2 (en) 2020-08-17 2024-03-05 ecoATM, Inc. Evaluating an electronic device using optical character recognition
CN114663418A (zh) * 2022-04-06 2022-06-24 京东安联财产保险有限公司 图像处理的方法及装置、存储介质及电子设备
CN115511907A (zh) * 2022-11-24 2022-12-23 深圳市晶台股份有限公司 一种用于led屏幕的划痕检测方法
CN116452613A (zh) * 2023-06-14 2023-07-18 山东省国土空间生态修复中心(山东省地质灾害防治技术指导中心、山东省土地储备中心) 一种地质调查中裂缝轮廓提取方法
CN116452613B (zh) * 2023-06-14 2023-08-29 山东省国土空间生态修复中心(山东省地质灾害防治技术指导中心、山东省土地储备中心) 一种地质调查中裂缝轮廓提取方法

Also Published As

Publication number Publication date
CN111325716B (zh) 2023-09-01
CN111325716A (zh) 2020-06-23

Similar Documents

Publication Publication Date Title
WO2021147387A1 (zh) 屏幕划痕碎裂检测方法及设备
WO2021147386A1 (zh) 屏幕划痕碎裂检测方法及设备
WO2021147385A1 (zh) 屏幕检测方法及装置
US20200160040A1 (en) Three-dimensional living-body face detection method, face authentication recognition method, and apparatuses
AU2017261537B2 (en) Automated selection of keeper images from a burst photo captured set
US8849062B2 (en) Eye defect detection in international standards organization images
WO2017101626A1 (zh) 一种实现图像处理的方法及装置
CN103617432B (zh) 一种场景识别方法及装置
US8306262B2 (en) Face tracking method for electronic camera device
US10382712B1 (en) Automatic removal of lens flares from images
US11790499B2 (en) Certificate image extraction method and terminal device
CN111325717B (zh) 手机缺陷位置识别方法及设备
CN111311556A (zh) 手机缺陷位置识别方法及设备
CN114283156B (zh) 一种用于去除文档图像颜色及手写笔迹的方法及装置
WO2021180131A1 (zh) 一种图像处理方法及电子设备
CN111626295A (zh) 车牌检测模型的训练方法和装置
CN113313626A (zh) 图像处理方法、装置、电子设备及存储介质
JP5640622B2 (ja) 赤目オブジェクト候補を分類する方法、コンピューター読み取り可能な媒体および画像処理装置
US8442348B2 (en) Image noise reduction for digital images using Gaussian blurring
WO2021147383A1 (zh) 相机模糊检测方法及装置
US20080199073A1 (en) Red eye detection in digital images
JP2007334876A (ja) 文書イメージ処理システムおよび文書イメージ処理方法
CN112183158B (zh) 一种谷物烹饪设备的谷物种类识别方法和谷物烹饪设备
WO2023225774A1 (zh) 图像处理方法及装置、电子设备、计算机可读存储介质
Long et al. An Efficient Method For Dark License Plate Detection

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: 20915454

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: 20915454

Country of ref document: EP

Kind code of ref document: A1