WO2021082923A1 - 一种电子设备屏幕区域瑕疵检测方法与设备 - Google Patents

一种电子设备屏幕区域瑕疵检测方法与设备 Download PDF

Info

Publication number
WO2021082923A1
WO2021082923A1 PCT/CN2020/120881 CN2020120881W WO2021082923A1 WO 2021082923 A1 WO2021082923 A1 WO 2021082923A1 CN 2020120881 W CN2020120881 W CN 2020120881W WO 2021082923 A1 WO2021082923 A1 WO 2021082923A1
Authority
WO
WIPO (PCT)
Prior art keywords
screen area
electronic device
image
defect
backbone network
Prior art date
Application number
PCT/CN2020/120881
Other languages
English (en)
French (fr)
Inventor
徐鹏
沈圣远
常树林
姚巨虎
Original Assignee
上海悦易网络信息技术有限公司
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 上海悦易网络信息技术有限公司 filed Critical 上海悦易网络信息技术有限公司
Priority to JP2022502026A priority Critical patent/JP2022539910A/ja
Publication of WO2021082923A1 publication Critical patent/WO2021082923A1/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
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • 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
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30121—CRT, LCD or plasma display

Definitions

  • This application relates to the field of computer technology, and in particular to a method and equipment for detecting defects in the screen area of an electronic device.
  • the traditional image processing method is based on the selection of the threshold to a large extent, and the screen area of second-hand electronic equipment such as mobile phones has different degrees of difference in various aspects such as color, appearance, aging, etc., it is difficult to give The determined threshold is therefore not applicable to the detection of defects in this screen area based on traditional image processing methods.
  • the purpose of this application is to provide a method and device for detecting defects in the screen area of an electronic device.
  • a method for detecting defects in a screen area of an electronic device including:
  • the defect detection result of the screen area image of the electronic device is received from the FPN network combined with the backbone network model.
  • the defect detection result includes: the defect type of the screen area of the electronic device and the position of the defect in the screen area of the electronic device And the confidence level of the defect detection result.
  • the extracting the screen area image in the appearance image includes:
  • the extracting the screen area of the electronic device includes:
  • the smallest bounding rectangle of the screen area is calculated to extract the screen area.
  • the clustering of all pixel values on the image based on the picture color histogram, and determining multiple clustering regions includes:
  • the cluster area is determined.
  • the cluster center is determined by the median of all pixels in the color histogram.
  • the judging the neighborhood of each clustering area and the connected domain with the largest statistics as the screen area of the electronic device includes:
  • the area with the largest connected domain is determined as the screen area of the electronic device.
  • the first two layers of the backbone network adopt a res structure
  • the last two layers of the network adopt an inception structure
  • the method further includes:
  • output result information including the defect type of the screen area of the electronic device and the position of the defect in the screen area of the electronic device.
  • the method further includes:
  • Step one preset the FPN network combined with the backbone network model and its initial model parameters
  • Step 2 Input the screen area image of the sample electronic device into the FPN network with current model parameters combined with the backbone network model to obtain the defect prediction result of the screen area of the sample electronic device.
  • the defect prediction result includes: The type of flaws in the screen area, the position of the flaws in the screen area of the sample electronic device, and the confidence level of the flaw detection results;
  • Step 3 Calculate the difference between the defect prediction result and the actual defect result of the screen area of the sample electronic device based on a preset objective function, and identify whether the difference is greater than a second preset threshold,
  • step 4 after updating the model parameters of the FPN network combined with the backbone network based on the difference, restart execution from step 2;
  • step 5 the FPN network combined with the backbone network model with the current model parameters is used as the model of the FPN network combined with the backbone network after the training.
  • the normalized parameters are updated iteratively after each convolution calculation is completed.
  • an electronic device screen area defect detection device which includes:
  • the memory is arranged to store computer-executable instructions, which when executed, cause the processor to: obtain the appearance image of the electronic device containing the screen area; extract the screen area image in the appearance image; The model of the FPN network combined with the backbone network after the regional image input training is completed; the defect detection result of the screen area image of the electronic device is received from the model of the FPN network combined with the backbone network, and the defect detection result includes: the screen of the electronic device The defect type of the area, the position of the defect in the screen area of the electronic device, and the confidence level of the defect detection result.
  • the processor Stored thereon are computer-readable instructions, where when the computer-executable instructions are executed by the processor, the processor: obtains the appearance image of the electronic device containing the screen area; extracts the screen area image in the appearance image; The model of the FPN network combined with the backbone network after the regional image input training is completed; the defect detection result of the screen area image of the electronic device is received from the model of the FPN network combined with the backbone network, and the defect detection result includes: the screen of the electronic device The defect type of the area, the position of the defect in the screen area of the electronic device, and the confidence level of the defect detection result.
  • this application obtains the appearance image of the electronic device including the screen area; extracts the screen area image in the appearance image; inputs the screen area image into the model of the FPN network combined with the backbone network after the training;
  • the FPN network combined with the backbone network model receives the output defect detection result of the screen area image of the electronic device, and the defect detection result includes: the defect type of the screen area of the electronic device, the position of the defect in the screen area of the electronic device, and
  • the confidence level of the defect detection result can accurately identify the difference in defects in the screen area of second-hand electronic devices such as mobile phones.
  • Fig. 1 shows a flow chart of a method for detecting defects in the screen area of an electronic device according to an aspect of the present application
  • FIG. 2 shows a schematic diagram of a screen area defect detection result according to an embodiment of the present invention
  • FIG. 3 shows a schematic diagram of a model of an FPN network combined with a backbone network according to an embodiment of the present invention.
  • each module and trusted party of the system includes 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), programmable read-only memory (PROM), Erase programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or flash memory (flash RAM).
  • RAM random access memory
  • ROM read-only memory
  • PROM programmable read-only memory
  • EPROM Erase programmable read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • flash RAM flash RAM
  • Memory is an example of computer readable media.
  • Computer-readable media include 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.
  • Fig. 1 shows a method for detecting defects in the screen area of an electronic device provided by one aspect of the present application, wherein the method includes:
  • S11 obtains the appearance image of the electronic device including the screen area
  • S13 inputs the screen area image into the FPN network after training and the model of the backbone network
  • the defect detection result of the screen area image of the electronic device is received from the model of the FPN network combined with the backbone network.
  • the defect detection result includes: the defect type of the screen area of the electronic device, and the defect in the screen area of the electronic device Confidence of location and defect detection results.
  • the method is executed by the device 1, which is a computer device and/or cloud.
  • the computer device includes, but is not limited to, a personal computer, a notebook computer, an industrial computer, a network host, a single network server, and multiple A set of network servers; the cloud is composed of a large number of computers or network servers based on Cloud Computing (Cloud Computing), where cloud computing is a type of distributed computing, a virtual supercomputer composed of a group of loosely coupled computer sets.
  • Cloud Computing Cloud Computing
  • the device 1 obtains the appearance image of the electronic device including the screen area.
  • the electronic device includes, but is not limited to, terminal devices such as mobile phones, PADs, smart watches, etc., wherein,
  • the screen area refers to the area where the display screen of the electronic device is located.
  • the method of this application can be used to identify the screen area of a second-hand mobile phone.
  • the device 1 can obtain the appearance image of the electronic device including the screen area by taking an image of the front of the electronic device, or the device 1 can receive the front image of the electronic device sent by the user equipment to realize the identification of the screen area.
  • the screen area image in the appearance image is extracted. Specifically, since the acquired appearance image includes the screen area and the non-screen area, the screen area needs to be identified and extracted to further detect the defects of the screen area.
  • step S12 includes:
  • S122 (not shown) clusters all pixel values on the image based on the picture color histogram, and determines multiple clustering regions;
  • S124 (not shown) extracts the screen area of the electronic device.
  • the picture color histogram of the appearance image is counted.
  • the picture color histogram of the appearance image including the screen area may be calculated based on an existing statistical method. statistics.
  • step S122 cluster all the pixel values on the image based on the picture color histogram to determine the cluster area.
  • the front appearance image of the electronic device usually includes a screen area and a non-screen area, two clustering areas are usually determined.
  • the step S122 includes: S1221 (not shown) determining the cluster center based on the picture color histogram; S1222 (not shown) based on the relationship between all pixel values on the image and the cluster center , Determine the clustering area.
  • the device 1 determines the cluster center based on the picture color histogram. Specifically, for example, the average value of all pixels may be determined through the picture color histogram, and the average value is used as the cluster center. Further, in the step S1222, the clustering area is determined by a clustering algorithm, for example, including but not limited to Kmeans clustering. Preferably, wherein the cluster center is determined by the median of all pixels in the color histogram.
  • the neighborhood of each clustering area is determined, and the connected domain with the largest statistics is used as the screen area of the electronic device.
  • the neighborhood of each cluster area includes, but is not limited to, 4 neighborhoods or 8 fields, and so on.
  • the step S123 includes: judging the 8 neighborhoods of each area, determining neighborhoods smaller than the pixel threshold as the same connected domain; and determining the area with the largest connected domain as the screen area of the electronic device.
  • the pixel threshold may be set in advance, or obtained through statistics, which is not limited here.
  • the screen area of the electronic device is extracted.
  • the screen area can be extracted, for example, the screen area part can be cut out.
  • the step S124 includes: calculating the minimum bounding rectangle of the screen area based on an opencv implementation method to extract the screen area.
  • the screen area can be extracted in the form of the smallest circumscribed rectangle for corresponding processing on the screen area.
  • the area corresponding to the smallest circumscribed rectangle is the screen area.
  • the screen area image of the electronic device is input into the model of the FPN network combined with the backbone network after the training.
  • the defect detection result of the screen area image of the electronic device is received from the model of the FPN network combined with the backbone network.
  • the defect detection result includes: the defect type of the screen area of the electronic device and the defect in the electronic device.
  • the position in the screen area of the device and the confidence level of the defect detection result include, but are not limited to, delamination, penetrating, leaking, broken lines, bright spots (bright spots), stains (yellow and blue), and so on.
  • the model of the FPN network combined with the backbone network can be shown in FIG. 3.
  • the FPN network combined with the backbone network model iteratively updates the normalized parameters after each convolution calculation is completed, where the normalized parameters include the mean and variance, and normalization ensures that each input is reasonable Changes within the range, here, the normalized parameters are continuously updated with the update of the input data.
  • each defect detection result includes cls, x1, y1, x2, y2, score, Among them, cls is the defect type, x1, y1, x2, y2 are the 4 coordinates of the position of the defect in the image of the screen area area, and score is the confidence level of this defect.
  • the present invention mainly utilizes the improved feature pyramid (FPN) network combined with the deep learning model of the backbone network, and can accurately identify the difference in the defects of the screen area of the second-hand electronic equipment such as the mobile phone.
  • FPN improved feature pyramid
  • the first two layers of the backbone network adopt a res structure
  • the last two layers of the network adopt an inception structure
  • step S14 after receiving the output defect detection result of the screen area of the electronic device from the FPN network combined with the backbone network model, further includes:
  • output result information including the defect type of the screen of the electronic device and the position of the defect in the screen area of the electronic device.
  • the method before inputting the screen area image into the FPN network combined with the backbone network model, the method further includes:
  • Step one preset the FPN network combined with the backbone network model and its initial model parameters
  • Step 2 Input the screen area image of the sample electronic device into the FPN network with current model parameters combined with the backbone network model to obtain the defect prediction result of the screen area of the sample electronic device.
  • the defect prediction result includes: The type of flaws in the screen area, the position of the flaws in the screen area of the sample electronic device, and the confidence level of the flaw detection results;
  • Step 3 Calculate the difference between the defect prediction result and the real defect result of the sample electronic device based on a preset objective function, and identify whether the difference is greater than a second preset threshold,
  • step 4 after updating the model parameters of the FPN network combined with the backbone network based on the difference, restart execution from step 2;
  • step 5 the FPN network combined with the backbone network model with the current model parameters is used as the model of the FPN network combined with the backbone network after the training.
  • the model of the FPN network combined with the backbone network is cyclically trained to obtain a reliable model.
  • an electronic device screen area defect detection device which includes:
  • the memory is arranged to store computer-executable instructions, which when executed, cause the processor to: obtain the appearance image of the electronic device containing the screen area; extract the screen area image in the appearance image; The model of the FPN network combined with the backbone network after the regional image input training is completed; the defect detection result of the screen area image of the electronic device is received from the model of the FPN network combined with the backbone network, and the defect detection result includes: the screen of the electronic device The defect type of the area, the position of the defect in the screen area of the electronic device, and the confidence level of the defect detection result.
  • the processor Stored thereon are computer-readable instructions, where when the computer-executable instructions are executed by the processor, the processor: obtains the appearance image of the electronic device containing the screen area; extracts the screen area image in the appearance image; The model of the FPN network combined with the backbone network after the regional image input training is completed; the defect detection result of the screen area image of the electronic device is received from the model of the FPN network combined with the backbone network, and the defect detection result includes: the screen of the electronic device The defect type of the area, the position of the defect in the screen area of the electronic device, and the confidence level of the defect detection result.
  • this application obtains the appearance image of the electronic device including the screen area; extracts the screen area image in the appearance image; inputs the screen area image into the model of the FPN network combined with the backbone network after the training;
  • the FPN network combined with the backbone network model receives the output of the defect detection result of the screen area image of the electronic device, and the defect detection result includes: the defect type of the screen area of the electronic device, the position of the defect in the screen area of the electronic device, and
  • the confidence level of the defect detection result can accurately identify the difference in defects in the screen area of second-hand electronic devices such as mobile phones.

Abstract

本申请的目的是提供一种电子设备屏幕区域瑕疵检测方法与设备。与现有技术相比，本申请通过获取电子设备的包含屏幕区域的外观图像；提取所述外观图像中的屏幕区域图像；将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度，能够准确地识别二手电子设备如手机的屏幕区域的瑕疵差异。

Description

一种电子设备屏幕区域瑕疵检测方法与设备 技术领域
本申请涉及计算机技术领域，尤其涉及一种电子设备屏幕区域瑕疵检测方法与设备。
背景技术
由于基于传统图像处理方式在很大程度上依赖于阈值的选取，而二手电子设备如手机等的屏幕区域由于在成色、外观、老化程度等各个方面都有不同程度的差异，故很难给出确定的阈值，因此基于传统图像处理方式的在本屏幕区域瑕疵检测中不适用。
发明内容
本申请的目的是提供一种电子设备屏幕区域瑕疵检测方法与设备。
根据本申请的一个方面，提供了一种电子设备屏幕区域瑕疵检测方法，所述方法包括：
获取电子设备的包含屏幕区域的外观图像；
提取所述外观图像中的屏幕区域图像；
将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；
从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。
进一步地，其特征在于，所述提取所述外观图像中的屏幕区域图像包括：
统计所述外观图像的图片颜色直方图；
基于所述图片颜色直方图对图像上的所有像素值做聚类，确定多个聚类区域；
判断每个聚类区域的邻域，统计最大的连通域作为所述电子设备的屏幕区域；
提取所述电子设备的屏幕区域。
进一步地，所述提取所述电子设备的屏幕区域包括：
基于opencv实现的方式，计算所述屏幕区域的最小外接矩形，以提取所述屏幕区域。
进一步地，其特征在于，所述基于所述图片颜色直方图对图像上的所有像素值做聚类，确定多个聚类区域包括：
基于所述图片颜色直方图确定聚类中心；
基于图像上的所有像素值与所述聚类中心的关系，确定聚类区域。
进一步地，所述聚类中心通过颜色直方图中所有像素的中位数来确定。
进一步地，其特征在于，所述判断每个聚类区域的邻域，统计最大的连通域作为所述电子设备的屏幕区域包括：
判断每个区域的8邻域，将小于像素阈值的邻域确定为同一连通域；
将具有最大连通域的区域确定为所述电子设备的屏幕区域。
进一步地，其中，所述backbone网络的前2层采用res结构，网络的后2层采用inception结构。
进一步地，其中，从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域的瑕疵检测结果之后，还包括：
识别所述瑕疵检测结果的置信度是否大于第一预设阈值，
若大于所述第一预设阈值，则输出包括电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置的结果信息。
进一步地，其中，将所述电子设备的屏幕区域图像输入FPN网络结合backbone网络的模型之前，还包括：
步骤一，预设FPN网络结合backbone网络的模型及其初始的模型参数；
步骤二，将样本电子设备的屏幕区域图像输入带有当前的模型参数的FPN网络结合backbone网络的模型，得到样本电子设备的屏幕区域的瑕疵预测结果，所述瑕疵预测结果包括：样本电子设备的屏幕区域的瑕疵种类、瑕疵在样本电子设备的屏幕区域中的位置和瑕疵检测结果的置信度；
步骤三，基于预设目标函数计算所述瑕疵预测结果与样本电子设备的屏幕区域的真实瑕疵结果之间的差值，识别所述差值是否大于第二预设阈，
若所述差值大于第二预设阈值，则步骤四，基于所述差值更新所述FPN网络结合backbone网络的模型参数后，重新从步骤二开始执行；
若所述差值小于等于第二预设阈值，则步骤五，将带有当前的模型参数的FPN网络结合backbone网络的模型作为训练结束后的FPN网络结合backbone网络的模型。
进一步地，其中，所述FPN网络结合backbone网络的模型中每一次卷积计算完成后迭代更新归一化的参数。
根据本申请的另一方面，还提供了一种电子设备屏幕区域瑕疵检测设备，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：获取电子设备的包含屏幕区域的外观图像；提取所述外观图像中的屏幕区域图像；将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。
根据本申请的再一方面，还提供了一种计算机可读介质，其中，
其上存储有计算机可读指令，其中，该计算机可执行指令被处理器执行时使得该处理器：获取电子设备的包含屏幕区域的外观图像；提取所述外观图像中的屏幕区域图像；将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。
与现有技术相比，本申请通过获取电子设备的包含屏幕区域的外观图像；提取所述外观图像中的屏幕区域图像；将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度，能够准确地识别二手电子设备如手机的屏幕 区域的瑕疵差异。
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示出根据本申请一个方面的一种电子设备屏幕区域瑕疵检测的方法流程图；
图2示出本发明一实施例的屏幕区域瑕疵检测结果的示意图；
图3示出本发明一实施例的FPN网络结合backbone网络的模型的示意图。
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，系统各模块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)、可编程只读存储器(PROM)、可擦除可编程只读存储器(EPROM)、电可擦除可编程只读存储器(EEPROM)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。
为更进一步阐述本申请所采取的技术手段及取得的效果，下面结合附图及优选实施例，对本申请的技术方案，进行清楚和完整的描述。
图1示出本申请一个方面提供的一种用于电子设备屏幕区域瑕疵检测方法，其中，该方法包括：
S11获取电子设备的包含屏幕区域的外观图像；
S12提取所述外观图像中的屏幕区域图像；
S13将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；
S14从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。
在本申请中，所述方法通过设备1执行，所述设备1为计算机设备和/或云，所述计算机设备包括但不限于个人计算机、笔记本电脑、工业计算机、网络主机、单个网络服务器、多个网络服务器集；所述云由基于云计算(Cloud Computing)的大量计算机或网络服务器构成，其中，云计算是分布式计算的一种，由一群松散耦合的计算机集组成的一个虚拟超级计算机。
在此，所述计算机设备和/或云仅为举例，其他现有的或者今后可能出现的设备和/或资源共享平台如适用于本申请也应包含在本申请的保护范围内，在此，以引用的方式包含于此。
在该实施例中，在所述步骤S11中，设备1获取电子设备的包含屏幕区域的外观图像，在此，所述电子设备包括但不限于手机、PAD、智能手表等等终端设备，其中，屏幕区域是指电子设备的显示屏所在区域。例如，本申请的方法可以用于二手手机的屏幕区域的识别。
具体地，设备1可通过拍摄电子设备的正面的图像来获取电子设备的包含屏幕区域的外观图像，或者，设备1可以接收用户设备发送的电子设 备的正面图像来实现对屏幕区域的识别。
继续在该实施例中，在所述步骤S12中，提取所述外观图像中的屏幕区域图像。具体地，由于获取到的外观图像包含屏幕区域及非屏幕区域，需要对屏幕区域进行识别并提取，以对屏幕区域的瑕疵做进一步检测。
优选地，其中，所述步骤S12包括：
S121(未示出)统计所述外观图像的图片颜色直方图；
S122(未示出)基于所述图片颜色直方图对图像上的所有像素值做聚类，确定多个聚类区域；
S123(未示出)判断每个聚类区域的邻域，统计最大的连通域作为所述电子设备的屏幕区域；
S124(未示出)提取所述电子设备的屏幕区域。
在该实施例中，在所述步骤S121中，统计所述外观图像的图片颜色直方图，具体地，可以基于现有的统计方式，来实现对包含屏幕区域的外观图像的图片颜色直方图的统计。
继续在该实施例中，在所述步骤S122中，基于所述图片颜色直方图对图像上的所有像素值做聚类，确定聚类区域。在此，由于电子设备的正面外观图像通常包含屏幕区域及非屏幕区域，因此，通常会确定出两个聚类区域。
优选地，其中，所述步骤S122包括：S1221(未示出)基于所述图片颜色直方图确定聚类中心；S1222(未示出)基于图像上的所有像素值与所述聚类中心的关系，确定聚类区域。
具体地，在所述步骤S1221中，设备1基于所述图片颜色直方图确定聚类中心，具体地，例如，可以通过图片颜色直方图确定出所有像素的均值，并将该均值作为聚类中心，进一步地，在所述步骤S1222中，通过聚类算法来确定聚类区域，例如，包括但不限于通过Kmeans聚类来实现。优选地，其中，所述聚类中心通过颜色直方图中所有像素的中位数来确定。
继续在该实施例中，在所述步骤S123中，判断每个聚类区域的邻域，统计最大的连通域作为所述电子设备的屏幕区域。在此，每个聚类区域的邻域包括但不限于4邻域或者8领域等等。
优选地，其中，所述步骤S123包括：判断每个区域的8邻域，将小于像 素阈值的邻域确定为同一连通域；将具有最大连通域的区域确定为所述电子设备的屏幕区域。在此，所述像素阈值可以提前设定好，或者经过统计得出，在此，不做限定。
继续在该实施例中，在所述步骤S124中，提取所述电子设备的屏幕区域。当识别出电子设备的屏幕区域后，可以对该屏幕区域进行提取，例如，截取出该屏幕区域部分。
优选地，其中，所述步骤S124包括：基于opencv实现的方式，计算所述屏幕区域的最小外接矩形，以提取所述屏幕区域。
在该实施例中，可以通过最小外接矩形的方式提取屏幕区域，以用于针对屏幕区域做相应处理。在此，最小外接矩形对应的区域为屏幕区域。
继续在该实施例中，在所述步骤S13中，将电子设备的屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型。
在所述步骤S14中，从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。其中，瑕疵种类包括但不限于分层、透字、漏液、断线、亮点(亮斑)、色斑(发黄发青)等等。
所述FPN网络结合backbone网络的模型可如图3所示。其中，所述FPN网络结合backbone网络的模型中每一次卷积计算完成后迭代更新归一化的参数，其中，归一化参数包括均值和方差，通过归一化确保每次输入都在合理的范围内变化，在此，所述归一化的参数随着输入数据的更新不断更新。
在此，从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域的瑕疵检测结果，如图2所示，每个瑕疵检测结果包含cls，x1，y1，x2，y2，score，其中，cls是缺陷类型，x1，y1，x2，y2是屏幕区域区域图像中瑕疵所在位置的4个坐标，score为这个瑕疵的置信度。
本发明主要利用改进的特征金字塔(FPN)网络结合backbone网络的深度学习模型，能够准确地识别二手电子设备如手机的屏幕区域的瑕疵差异。
本发明的屏幕区域瑕疵检测方法一实施例中，所述backbone网络的前2层采用res结构，网络的后2层采用inception结构。
本发明的屏幕区域瑕疵检测方法一实施例中，步骤S14，从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域的瑕疵检测结果之后，还包括：
识别所述瑕疵检测结果的置信度是否大于第一预设阈值，
若大于所述第一预设阈值，则输出包括电子设备的屏幕的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置的结果信息。
本实施例通过识别所述瑕疵检测结果的置信度，可以从瑕疵检测结果中筛选出可靠的结果进行输出。
本发明的屏幕区域瑕疵检测方法一实施例中，将所述屏幕区域图像输入FPN网络结合backbone网络的模型之前，还包括：
步骤一，预设FPN网络结合backbone网络的模型及其初始的模型参数；
步骤二，将样本电子设备的屏幕区域图像输入带有当前的模型参数的FPN网络结合backbone网络的模型，得到样本电子设备的屏幕区域的瑕疵预测结果，所述瑕疵预测结果包括：样本电子设备的屏幕区域的瑕疵种类、瑕疵在样本电子设备的屏幕区域中的位置和瑕疵检测结果的置信度；
步骤三，基于预设目标函数计算所述瑕疵预测结果与样本电子设备的真实瑕疵结果之间的差值，识别所述差值是否大于第二预设阈，
若所述差值大于第二预设阈值，则步骤四，基于所述差值更新所述FPN网络结合backbone网络的模型参数后，重新从步骤二开始执行；
若所述差值小于等于第二预设阈值，则步骤五，将带有当前的模型参数的FPN网络结合backbone网络的模型作为训练结束后的FPN网络结合backbone网络的模型。
在此，通过识别所述差值是否大于第二预设阈，来循环训练FPN网络结合backbone网络的模型，能够得到可靠的模型。
根据本申请的另一方面，还提供了一种电子设备屏幕区域瑕疵检测设备，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：获取电子设备的包含屏幕区域的外观图像；提取所述外观图 像中的屏幕区域图像；将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。
根据本申请的再一方面，还提供了一种计算机可读介质，其中，
其上存储有计算机可读指令，其中，该计算机可执行指令被处理器执行时使得该处理器：获取电子设备的包含屏幕区域的外观图像；提取所述外观图像中的屏幕区域图像；将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。
与现有技术相比，本申请通过获取电子设备的包含屏幕区域的外观图像；提取所述外观图像中的屏幕区域图像；将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度，能够准确地识别二手电子设备如手机的屏幕区域的瑕疵差异。
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 (12)

  1. 一种电子设备屏幕区域瑕疵检测方法，所述方法包括：
    获取电子设备的包含屏幕区域的外观图像；
    提取所述外观图像中的屏幕区域图像；
    将屏幕区域图像输入训练结束后的FPN网络结合backbone网络的模型；
    从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域图像的瑕疵检测结果，所述瑕疵检测结果包括：电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置和瑕疵检测结果的置信度。
  2. 根据权利要求1所述的方法，其特征在于，所述提取所述外观图像中的屏幕区域图像包括：
    统计所述外观图像的图片颜色直方图；
    基于所述图片颜色直方图对图像上的所有像素值做聚类，确定多个聚类区域；
    判断每个聚类区域的邻域，统计最大的连通域作为所述电子设备的屏幕区域；
    提取所述电子设备的屏幕区域。
  3. 根据权利要求2所述的方法，其特征在于，所述提取所述电子设备的屏幕区域包括：
    基于opencv实现的方式，计算所述屏幕区域的最小外接矩形，以提取所述屏幕区域。
  4. 根据权利要求2所述的方法，其特征在于，所述基于所述图片颜色直方图对图像上的所有像素值做聚类，确定多个聚类区域包括：
    基于所述图片颜色直方图确定聚类中心；
    基于图像上的所有像素值与所述聚类中心的关系，确定聚类区域。
  5. 根据权利要求4所述的方法，其特征在于，所述聚类中心通过颜色直方图中所有像素的中位数来确定。
  6. 根据权利要求2所述的方法，其特征在于，所述判断每个聚类区域的邻域，统计最大的连通域作为所述电子设备的屏幕区域包括：
    判断每个区域的8邻域，将小于像素阈值的邻域确定为同一连通域；
    将具有最大连通域的区域确定为所述电子设备的屏幕区域。
  7. 根据权利要求1所述的方法，其中，所述backbone网络的前2层采用res结构，网络的后2层采用inception结构。
  8. 根据权利要求1所述的方法，其中，从所述FPN网络结合backbone网络的模型接收输出的电子设备的屏幕区域的瑕疵检测结果之后，还包括：
    识别所述瑕疵检测结果的置信度是否大于第一预设阈值，
    若大于所述第一预设阈值，则输出包括电子设备的屏幕区域的瑕疵种类、瑕疵在电子设备的屏幕区域中的位置的结果信息。
  9. 根据权利要求1所述的方法，其中，将所述电子设备的屏幕区域图像输入FPN网络结合backbone网络的模型之前，还包括：
    步骤一，预设FPN网络结合backbone网络的模型及其初始的模型参数；
    步骤二，将样本电子设备的屏幕区域图像输入带有当前的模型参数的FPN网络结合backbone网络的模型，得到样本电子设备的屏幕区域的瑕疵预测结果，所述瑕疵预测结果包括：样本电子设备的屏幕区域的瑕疵种类、瑕疵在样本电子设备的屏幕区域中的位置和瑕疵检测结果的置信度；
    步骤三，基于预设目标函数计算所述瑕疵预测结果与样本电子设备的屏幕区域的真实瑕疵结果之间的差值，识别所述差值是否大于第二预设阈，
    若所述差值大于第二预设阈值，则步骤四，基于所述差值更新所述FPN网络结合backbone网络的模型参数后，重新从步骤二开始执行；
    若所述差值小于等于第二预设阈值，则步骤五，将带有当前的模型参数的FPN网络结合backbone网络的模型作为训练结束后的FPN网络结合backbone网络的模型。
  10. 根据权利要求1所述的方法，其中，所述FPN网络结合backbone网络的模型中每一次卷积计算完成后迭代更新归一化的参数。
  11. 一种计算机可读介质，其中，
    其上存储有计算机可读指令，所述计算机可读指令可被处理器执行以实现如权利要求1至10任一项所述的方法。
  12. 一种用于电子设备屏幕区域识别的设备，其中，所述设备包括：
    一个或多个处理器；以及
    存储有计算机可读指令的存储器，所述计算机可读指令在被执行时使所 述处理器执行如权利要求1至10中任一项所述方法的操作。
PCT/CN2020/120881 2019-10-28 2020-10-14 一种电子设备屏幕区域瑕疵检测方法与设备 WO2021082923A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2022502026A JP2022539910A (ja) 2019-10-28 2020-10-14 電子デバイススクリーンエリア瑕疵検査方法及び装置

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911032879.7A CN110796646A (zh) 2019-10-28 2019-10-28 一种电子设备屏幕区域瑕疵检测方法与设备
CN201911032879.7 2019-10-28

Publications (1)

Publication Number Publication Date
WO2021082923A1 true WO2021082923A1 (zh) 2021-05-06

Family

ID=69441639

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/120881 WO2021082923A1 (zh) 2019-10-28 2020-10-14 一种电子设备屏幕区域瑕疵检测方法与设备

Country Status (3)

Country Link
JP (1) JP2022539910A (zh)
CN (1) CN110796646A (zh)
WO (1) WO2021082923A1 (zh)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113569737A (zh) * 2021-07-28 2021-10-29 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) 基于自主学习网络模型笔记本屏幕缺陷检测方法及介质
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
US11798250B2 (en) 2019-02-18 2023-10-24 Ecoatm, Llc Neural network based physical condition evaluation of electronic devices, and associated systems and methods
CN117036354A (zh) * 2023-10-09 2023-11-10 深圳市魔样科技有限公司 一种智能戒指显示屏幕检测方法
CN117058411A (zh) * 2023-10-13 2023-11-14 东声(苏州)智能科技有限公司 电池棱边外形瑕疵识别方法、装置、介质和设备
US11922467B2 (en) 2020-08-17 2024-03-05 ecoATM, Inc. Evaluating an electronic device using optical character recognition
CN117890389A (zh) * 2024-03-14 2024-04-16 上海强华实业股份有限公司 精烧制品表面质量在线检测方法、设备及其存储介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN110796646A (zh) * 2019-10-28 2020-02-14 上海悦易网络信息技术有限公司 一种电子设备屏幕区域瑕疵检测方法与设备
CN110827249A (zh) * 2019-10-28 2020-02-21 上海悦易网络信息技术有限公司 电子设备背板外观瑕疵检测方法及设备
CN111680750B (zh) * 2020-06-09 2022-12-06 创新奇智(合肥)科技有限公司 图像识别方法、装置和设备

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107123111A (zh) * 2017-04-14 2017-09-01 浙江大学 一种用于手机屏幕缺陷检测的深度残差网络构造方法
EP3115935B1 (en) * 2015-06-29 2019-01-30 Nokia Technologies Oy A method, apparatus, computer program and system for image analysis
CN109859163A (zh) * 2018-12-19 2019-06-07 重庆邮电大学 一种基于特征金字塔卷积神经网络的lcd缺陷检测方法
CN109978875A (zh) * 2019-04-03 2019-07-05 无锡立赫智能科技有限公司 一种电容器外观缺陷识别方法及识别装置
CN110349135A (zh) * 2019-06-27 2019-10-18 歌尔股份有限公司 目标检测方法和装置
CN110796647A (zh) * 2019-10-28 2020-02-14 上海悦易网络信息技术有限公司 一种电子设备屏幕区域瑕疵检测方法与设备
CN110796646A (zh) * 2019-10-28 2020-02-14 上海悦易网络信息技术有限公司 一种电子设备屏幕区域瑕疵检测方法与设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004005383A (ja) * 2002-04-19 2004-01-08 Sony Corp 画像処理方法、画像処理装置、プログラム及び記録媒体、自動トリミング装置並びに肖像写真撮影装置
CN105894036B (zh) * 2016-04-19 2019-04-09 武汉大学 一种应用于手机屏幕缺陷检测的图像特征模板匹配方法
US11210777B2 (en) * 2016-04-28 2021-12-28 Blancco Technology Group IP Oy System and method for detection of mobile device fault conditions
CN109711474B (zh) * 2018-12-24 2023-01-17 中山大学 一种基于深度学习的铝材表面缺陷检测算法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3115935B1 (en) * 2015-06-29 2019-01-30 Nokia Technologies Oy A method, apparatus, computer program and system for image analysis
CN107123111A (zh) * 2017-04-14 2017-09-01 浙江大学 一种用于手机屏幕缺陷检测的深度残差网络构造方法
CN109859163A (zh) * 2018-12-19 2019-06-07 重庆邮电大学 一种基于特征金字塔卷积神经网络的lcd缺陷检测方法
CN109978875A (zh) * 2019-04-03 2019-07-05 无锡立赫智能科技有限公司 一种电容器外观缺陷识别方法及识别装置
CN110349135A (zh) * 2019-06-27 2019-10-18 歌尔股份有限公司 目标检测方法和装置
CN110796647A (zh) * 2019-10-28 2020-02-14 上海悦易网络信息技术有限公司 一种电子设备屏幕区域瑕疵检测方法与设备
CN110796646A (zh) * 2019-10-28 2020-02-14 上海悦易网络信息技术有限公司 一种电子设备屏幕区域瑕疵检测方法与设备

Cited By (12)

* 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
CN113569737A (zh) * 2021-07-28 2021-10-29 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) 基于自主学习网络模型笔记本屏幕缺陷检测方法及介质
CN117036354A (zh) * 2023-10-09 2023-11-10 深圳市魔样科技有限公司 一种智能戒指显示屏幕检测方法
CN117036354B (zh) * 2023-10-09 2024-02-02 深圳市魔样科技有限公司 一种智能戒指显示屏幕检测方法
CN117058411A (zh) * 2023-10-13 2023-11-14 东声(苏州)智能科技有限公司 电池棱边外形瑕疵识别方法、装置、介质和设备
CN117058411B (zh) * 2023-10-13 2024-04-09 东声(苏州)智能科技有限公司 电池棱边外形瑕疵识别方法、装置、介质和设备
CN117890389A (zh) * 2024-03-14 2024-04-16 上海强华实业股份有限公司 精烧制品表面质量在线检测方法、设备及其存储介质

Also Published As

Publication number Publication date
JP2022539910A (ja) 2022-09-13
CN110796646A (zh) 2020-02-14

Similar Documents

Publication Publication Date Title
WO2021082923A1 (zh) 一种电子设备屏幕区域瑕疵检测方法与设备
WO2021082918A1 (zh) 屏幕外观瑕疵检测方法及设备
WO2021082921A1 (zh) 电子设备背板外观瑕疵检测方法及设备
WO2021082919A1 (zh) 一种电子设备屏幕区域瑕疵检测方法与设备
WO2021082920A1 (zh) 电子设备边框外观瑕疵检测方法及设备
WO2019218824A1 (zh) 一种移动轨迹获取方法及其设备、存储介质、终端
CN110827244A (zh) 一种用于电子设备外观瑕疵检测的方法与设备
US11594053B2 (en) Deep-learning-based identification card authenticity verification apparatus and method
US20190370551A1 (en) Object detection and tracking delay reduction in video analytics
CN112949767B (zh) 样本图像增量、图像检测模型训练及图像检测方法
CN111666907B (zh) 一种视频中对象信息的识别方法、装置及服务器
WO2021082922A1 (zh) 一种检测屏幕显示断线的方法及设备
WO2020253616A1 (zh) 音频采集设备定位方法及装置、说话人识别方法及系统
WO2021013049A1 (zh) 前景图像获取方法、前景图像获取装置和电子设备
CN110827248A (zh) 一种电子设备屏幕区域识别方法与设备
TWI765442B (zh) 瑕疵等級判定的方法及存儲介質
CN115861400B (zh) 目标对象检测方法、训练方法、装置以及电子设备
CN115063656A (zh) 图像检测方法、装置、计算机可读存储介质及电子设备
CN108932703B (zh) 图片处理方法、图片处理装置及终端设备
CN108763491B (zh) 图片处理方法、装置及终端设备
CN108764206B (zh) 目标图像识别方法和系统、计算机设备
CN108932704B (zh) 图片处理方法、图片处理装置及终端设备
US10631050B2 (en) Determining and correlating visual context on a user device with user behavior using digital content on the user device
CN115205555B (zh) 确定相似图像的方法、训练方法、信息确定方法及设备
CN117115568B (zh) 数据筛选方法、装置、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20881041

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022502026

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20881041

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20881041

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 26.01.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20881041

Country of ref document: EP

Kind code of ref document: A1