WO2019218459A1 - 一种照片存储方法、存储介质、服务器和装置 - Google Patents

一种照片存储方法、存储介质、服务器和装置 Download PDF

Info

Publication number
WO2019218459A1
WO2019218459A1 PCT/CN2018/097097 CN2018097097W WO2019218459A1 WO 2019218459 A1 WO2019218459 A1 WO 2019218459A1 CN 2018097097 W CN2018097097 W CN 2018097097W WO 2019218459 A1 WO2019218459 A1 WO 2019218459A1
Authority
WO
WIPO (PCT)
Prior art keywords
photo
value
preset
shooting
classification
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2018/097097
Other languages
English (en)
French (fr)
Inventor
乐志能
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
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 Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2019218459A1 publication Critical patent/WO2019218459A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • 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
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • the present application relates to the field of computer technologies, and in particular, to a photo storage method, a storage medium, a server, and a device.
  • smart terminals having a photographing function generally store photographs taken according to the shooting time, and some may also be classified according to the APP of the photo source, but they cannot be automatically classified and stored according to the information in the photographs.
  • the identification can be searched only by manual operation, thereby wasting a lot of manpower and time.
  • the embodiment of the present application provides a photo storage method, a storage medium, a server, and a device.
  • a photo storage method when a user needs to select a photo including the same photo information from a large number of photos, the manual operation can only be performed.
  • a lookup identification which wastes a lot of manpower and time issues.
  • a first aspect of the embodiments of the present application provides a photo storage method, including:
  • a storage path is automatically generated, the storage path corresponds to a new photo storage folder, and the photo is stored in the new photo storage folder.
  • a second aspect of an embodiment of the present application provides a server comprising a memory and a processor, the memory storing computer readable instructions executable on the processor, the processor executing the computer readable instructions The following steps are implemented:
  • a storage path is automatically generated, the storage path corresponds to a new photo storage folder, and the photo is stored in the new photo storage folder.
  • a third aspect of embodiments of the present application provides a computer readable storage medium storing computer readable instructions that, when executed by a processor, implement the following steps:
  • a storage path is automatically generated, the storage path corresponds to a new photo storage folder, and the photo is stored in the new photo storage folder.
  • a fourth aspect of the embodiments of the present application provides a photo storage device, including:
  • An attribute obtaining unit configured to acquire a shooting attribute of a photo
  • a categorization value calculation unit configured to calculate a classification attribution value of the photo according to a shooting attribute of the photo and a preset attribute weight
  • a first storage unit configured to search a photo storage folder corresponding to the classification attribution value in the preset classification comparison table, and store the photo in a photo storage folder corresponding to the classification attribution value;
  • a second storage unit configured to automatically generate a storage path if the photo storage folder corresponding to the classification attribution value is not found, the storage path corresponding to a new photo storage folder, and storing the photo in the storage location The new photo storage folder.
  • the photographing attribute of the photo by acquiring the photographing attribute of the photo, calculating the classification attribution value of the photo according to the photographing attribute of the photograph and the preset attribute weight, and then searching for the classified attribution value in the preset classification comparison table.
  • Corresponding photo storage folder and storing the photo in a photo storage folder corresponding to the classification attribution value, if a photo storage folder corresponding to the classification attribution value is not found, a storage path is automatically generated.
  • the storage path corresponds to a new photo storage folder, and the photo is stored in the new photo storage folder, and the photos can be classified and stored when the shooting is completed, without the user manually storing the photos, saving the user. Time and improve the efficiency of photo storage.
  • FIG. 1 is a flowchart of an implementation of a photo storage method provided by an embodiment of the present application
  • FIG. 2 is a flowchart of a specific implementation of a photo storage method S102 provided by an embodiment of the present application
  • FIG. 3 is a specific implementation flowchart of a photo storage method B2 provided by an embodiment of the present application.
  • FIG. 4 is a specific implementation flowchart of a photo storage method S103 provided by an embodiment of the present application.
  • FIG. 5 is a flowchart of an implementation of a photo storage method according to another embodiment of the present application.
  • FIG. 6 is a structural block diagram of a photo storage device according to an embodiment of the present application.
  • FIG. 7 is a structural block diagram of a photo storage device according to another embodiment of the present application.
  • FIG. 8 is a schematic diagram of a server provided by an embodiment of the present application.
  • FIG. 1 shows an implementation flow of a photo storage method provided by an embodiment of the present application, and the method flow includes steps S101 to S104.
  • the specific implementation principles of each step are as follows:
  • S101 Acquire a shooting attribute of a photo.
  • the shooting attributes include shooting time, shooting location, and photo type.
  • photo types include portraits, food photos, landscape photos, and two-dimensional codes.
  • the photographing attribute of the photo may be determined according to a user-defined label for taking a photo using the smart terminal. For example, at 10 o'clock on January 20, 2018, a self-portrait face shot is taken at home, where the location can be customized by user-defined location label, such as defining a location as a home, work unit, or an attraction. Users may take photos at the same location at the same time each day, for example, a QR code taken at the parking lot when they go to work.
  • the location information of the smart terminal is determined by the big data to determine the shooting location, for example, the day is divided according to the working time period and the rest time period, for example, 9:00 am to 6:00 pm is the working time period, and at night 11 Point to 6 am the next morning, for the rest period.
  • the geographic location of the smart terminal in the preset working day is clustered and analyzed. Specifically, the geographic location of all working time periods in the preset working day is stored in the working time set, and the first specified number of geographical locations are randomly selected as the first cluster center, and each geographical position in the working time set is calculated. The distance value from the first cluster center is initially clustered according to the calculated distance value and the specified cluster center, and the second place is selected from the geographical position after the initial clustering.
  • the location label that defines the cluster center (geographical location) of the set of working hours after convergence is the work unit.
  • the location label of the cluster center (geographical location) defining the set of working hours after convergence is home. It should be noted that the shooting location includes not only the work unit and the home, but also the parking lot, the school, etc., and other shooting locations can also be determined according to the above method.
  • the type of photo for the photo can be determined by entering the photo into a trained convolutional neural network model. Specifically, an image feature of the photo is extracted, an image feature of the extracted photo is input to an input layer of the trained convolutional neural network model, and a photo type of the photo is output at the output layer.
  • the trained convolutional neural network model is obtained according to the following steps:
  • A1 Acquire a set number of sample photos, which are pre-set with type tags. Specifically, by taking a set number of sample photos, the sample photos are pre-set with labels such as landscapes, people, food, two-dimensional codes, and the like.
  • A2 Establish a convolutional neural network model including an input layer, a convolutional layer, a fully connected layer, and an output layer.
  • A3 in the first training, presetting the network connection weight and the threshold between the nodes of the convolutional neural network model to a random value satisfying the preset condition, and setting an ideal output value of the sample photo, Selecting a sample photo from the set number of sample photos, inputting to the input layer, passing through the convolution layer and the fully connected layer, transmitting to the output layer, obtaining the actual output value of the sample photo, completing a round of training, and Calculate the difference between the actual output value and the ideal output value.
  • A4 According to the calculated difference, adjust the network connection weight and threshold between each layer node according to the specified learning rule, and then train the convolutional neural network model again until the calculated difference is not greater than the preset At the threshold, the training is completed and the trained convolutional neural network model is obtained.
  • a convolutional neural network model including an input layer, a convolution layer, a fully connected layer, and an output layer
  • the training points are as follows.
  • a sample input convolutional neural network model is randomly selected from the sample photos, and an output value of the sample photo is calculated.
  • the network connection weights and thresholds between the nodes of the convolutional neural network model are pre-set to a small random value close to 0, and the ideal output value of the sample photo is set.
  • the sample photo is transmitted from the input layer to the output layer through the convolution layer and the fully connected layer, and the actual output value of the sample photo is obtained, and a round of training is completed to calculate the difference between the actual output value and the ideal output value.
  • the global difference value D of the convolutional neural network is calculated according to the following formula:
  • n is a positive integer
  • n is the number of samples for the total number of pictures of the training.
  • Adjust the weight matrix according to the method of minimizing the error. Set the error threshold. If D is greater than the threshold, adjust the network connection weight and threshold between the nodes in the layer according to the Delta learning rule, and then train the convolutional neural network model again until the global error D of the network is not greater than the threshold. Up to the threshold, the training is ended, and the weight and threshold of the training are saved as the optimal model parameters of the convolutional neural network, and the trained convolutional neural network model is obtained.
  • the convolutional neural network model for training, determining an optimal model parameter of the neural network model, thereby obtaining a trained convolutional neural network model, The captured photos are input into the trained convolutional neural network model to quickly obtain the photo type of the photo, thereby improving the efficiency of the photo classification storage.
  • S102 Calculate a classification attribution value of the photo according to the shooting attribute of the photo and the preset attribute weight.
  • each shooting attribute corresponds to a preset attribute weight.
  • the shooting location corresponds to a preset location weight
  • the shooting time corresponds to a preset time weight
  • the photo type corresponds to a preset type weight.
  • the foregoing S102 specifically includes:
  • mapping according to the attribute weight corresponding to the shooting attribute, a shooting time, a shooting location, and a photo type of the photo into the three-dimensional coordinate system, and determining three-dimensional coordinates of the photo in the three-dimensional coordinate system . Specifically, the photographing attribute of the photo is mapped to a point in the three-dimensional coordinate system.
  • B3 Calculate a distance value of a point corresponding to the three-dimensional coordinate of the photo and an origin of the three-dimensional coordinate system, and use the distance value as a classification attribution value of the photo.
  • the photo is mapped to a point in the three-dimensional coordinate system, and the distance value of the photo corresponding to the origin in the three-dimensional coordinate system is used as the classification attribution value of the photo, and the photo is taken.
  • the attributes are quantified for classification storage based on the classification attribution values.
  • FIG. 3 shows a specific implementation process of the photo storage method step B2 provided by the embodiment of the present application, which is described in detail as follows:
  • a mapping table is set in advance, and the mapping table includes numerical values corresponding to different shooting times, numerical values corresponding to different shooting locations, and numerical values corresponding to different photo types. Specifically, the shooting time is divided according to the shooting date or the shooting time. In the mapping table, if the shooting time is divided according to the shooting date, the value corresponding to the shooting date is incremented by the specified difference; if the shooting time is pressed The shooting time is divided, and the shooting time is divided into several time segments, and the values corresponding to the time segments are incremented by the specified difference step by step. If the shooting time period is, for example, the value corresponding to 10 o'clock in the morning is 5, and the value corresponding to 3 o'clock in the afternoon is 8.
  • the mapping table includes values corresponding to different latitude and longitude.
  • the type of photographing is determined by predetermining the type of photo determined by statistical analysis of big data, and setting different values for different photo types, such as 10 for a person and 20 for a landscape.
  • B22 Find a value corresponding to the shooting time, the shooting location, and the photo type of the photo from the mapping table.
  • B23 Determine three-dimensional coordinates of the photo in the three-dimensional coordinate system according to respective values corresponding to the photographing time, the photographing place, and the photo type of the photograph and the attribute weight. Specifically, the product of the photographing time corresponding to the photographing time and the preset time weight is used as the value on the first coordinate axis in the three-dimensional coordinate system, and the numerical value corresponding to the photographing location of the photograph is preset. The product of the location weights is used as the value on the second coordinate axis in the three-dimensional coordinate system, and the product of the value corresponding to the photo type of the photo and the preset type weight is used as the third coordinate axis in the three-dimensional coordinate system. The value of the three-dimensional coordinates of the photo in the three-dimensional coordinate system.
  • S103 Search a photo storage folder corresponding to the classification attribution value in the preset classification comparison table, and store the photo in a photo storage folder corresponding to the classification attribution value.
  • the preset classification comparison table includes a correspondence between the classification attribution value and a preset numerical value corresponding to the photo storage folder.
  • FIG. 4 shows a specific implementation process of the photo storage method S103 provided by the embodiment of the present application, which is described in detail as follows:
  • C1 Obtain a preset value corresponding to each of the photo storage folders from the preset classification comparison table. Specifically, the preset value corresponding to the photo storage folder may be determined according to a classification attribution value stored in the first photo.
  • C2 Compare the classified attribution value of the photo with the obtained preset value one by one, and determine the preset value that has the smallest absolute value of the difference from the classified attribution value of the photo.
  • C3 determining, if the determined absolute value of the difference between the preset value and the classified attribution value of the photo is in a preset difference interval, determining the determined preset value and the classified attribution value of the photo match. Certainly, if the absolute value of the difference between the preset value and the classified attribution value of the photo is not within the preset difference interval, determining the determined preset value and the classification of the photo The values do not match.
  • C4 storing the photo in the photo storage folder corresponding to the determined preset value. Specifically, the photo is stored in a photo storage folder corresponding to a preset value matching the classification attribution value of the photo.
  • a difference between the preset value and the classified attribution value of the photo is defined as E, and when E ⁇ , the photo is stored in a photo storage folder corresponding to the preset value, when E When ⁇ , that is, the preset classification table does not have the absolute value of the difference value of the classification attribute value of the photo in the preset difference interval, and at this time, a storage is automatically generated.
  • Path the storage path corresponds to a new photo storage folder
  • the folder is said to be a file storage path, that is, by automatically generating a storage path, creating a new photo storage folder, and storing the photo in the newly created file In the folder.
  • the photo storage folder corresponding to the classification attribution value is not found, a storage path automatically generated corresponding to the existing photo storage folder is different. Further, the classification attribution value of the photo is used as a preset value of the photo storage file.
  • the photo is further classified according to the area value of the face in the photo, for example, further subdivided into a selfie.
  • the face recognition is performed on the person photo, and the area value of the recognized face is calculated, and the area ratio of the face area in the person photo is determined, and if the area ratio is greater than the preset area ratio threshold Then, the person photo is further stored in the selfie folder, so that the user can quickly process the selfie.
  • the similarity of the photos may be calculated, and the subfolders of the photo storage folder are created according to the similarity of the photos, and the photos with the similarity greater than the preset similarity threshold are Save in the same subfolder. For example, it is determined whether the similarity of the photos is the same face, and if so, a subfolder is created in the photo storage folder, and the photo of the same face is stored in the subfolder.
  • the photographing attribute of the photo by acquiring the photographing attribute of the photo, calculating the classification attribution value of the photo according to the photographing attribute of the photograph and the preset attribute weight, and then searching for the classified attribution value in the preset classification comparison table.
  • Corresponding photo storage folder and storing the photo in a photo storage folder corresponding to the classification attribution value, if a photo storage folder corresponding to the classification attribution value is not found, a storage path is automatically generated.
  • the storage path corresponds to a new photo storage folder, and the photo is stored in the new photo storage folder, and the photos can be classified and stored when the shooting is completed, without the user manually storing the photos, saving the user. Time and improve the efficiency of photo storage.
  • the photo storage method further includes:
  • S202 Calculate a classification attribution value of the photo according to the shooting attribute of the photo and the preset attribute weight.
  • S203 Search a photo storage folder corresponding to the classification attribution value in the preset classification comparison table, and store the photo in a photo storage folder corresponding to the classification attribution value.
  • step S201 to the step S204 For the specific steps of the step S201 to the step S204, refer to the steps S101 to S104 in the foregoing embodiment, and details are not described herein again.
  • S205 Arrange the photos stored in the same photo storage folder according to a timeline stored in the photo storage folder.
  • the shooting time has priority over the photo type
  • the photo type has a higher priority than the shooting location
  • the photos in the same photo storage folder are arranged according to the stored timeline, when the photo storage folder is newly saved.
  • the newly saved photo is replaced with the cover of the photo storage folder, and becomes a new cover of the photo storage folder, so that the cover of the photo storage folder is dynamically updated according to the stored photo, and the user does not need to open the file.
  • the folder quickly identifies the type of photos stored in the folder and improves the user experience.
  • the photographing attribute of the photo by acquiring the photographing attribute of the photo, calculating the classification attribution value of the photo according to the photographing attribute of the photograph and the preset attribute weight, and then searching for the classified attribution value in the preset classification comparison table.
  • Corresponding photo storage folder and storing the photo in a photo storage folder corresponding to the classification attribution value, if a photo storage folder corresponding to the classification attribution value is not found, a storage path is automatically generated.
  • the storage path corresponds to a new photo storage folder, and the photo is stored in the new photo storage folder, and the photos can be classified and stored when the shooting is completed, without the user manually storing the photos, saving the user.
  • the photos stored in the same photo storage folder are arranged according to the timeline stored in the photo storage folder, according to the time stored in the photo storage folder Line, the latest photo stored in the folder as the cover of the photo storage folder, dynamically update the cover, the user does not need to point Folders to quickly clear picture of the type of folder storage, improve the user experience.
  • FIG. 6 is a block diagram showing the structure of the photo storage device provided by the embodiment of the present application. For the convenience of description, only the parts related to the embodiment of the present application are shown.
  • the photo storage device includes: an attribute obtaining unit 61, a home value calculating unit 62, a first storage unit 63, and a second storage unit 64, wherein:
  • An attribute obtaining unit 61 configured to acquire a shooting attribute of a photo
  • the attribution value calculation unit 62 is configured to calculate a classification attribution value of the photo according to the photographing attribute of the photo and the preset attribute weight;
  • the first storage unit 63 is configured to search a photo storage folder corresponding to the classification attribution value in the preset classification comparison table, and store the photo in a photo storage folder corresponding to the classification attribution value;
  • the second storage unit 64 is configured to automatically generate a storage path if the photo storage folder corresponding to the classification attribution value is not found, the storage path corresponding to a new photo storage folder, and storing the photo The new photo storage folder.
  • the attribution value calculation unit 62 includes:
  • a coordinate system establishing module for establishing a three-dimensional coordinate system
  • a coordinate determining module configured to map a shooting time, a shooting location, and a photo type of the photo into the three-dimensional coordinate system according to the attribute weight corresponding to the shooting attribute, and determine that the photo is in the three-dimensional coordinate system Three-dimensional coordinates in ;
  • the attribution value determining module is configured to calculate a distance value between a point corresponding to the three-dimensional coordinate of the photo and an origin of the three-dimensional coordinate system, and use the distance value as a classification attribution value of the photo.
  • the coordinate determining module includes:
  • a preset sub-module configured to pre-establish a mapping table, where the mapping table includes values corresponding to different shooting times, values corresponding to different shooting locations, and values corresponding to different photo types;
  • a value finding sub-module configured to search, from the mapping table, a value corresponding to a shooting time, a shooting location, and a photo type of the photo;
  • a coordinate determining sub-module configured to determine a three-dimensional coordinate of the photo in the three-dimensional coordinate system according to a value corresponding to a shooting time, a shooting location, and a photo type of the photo and the attribute weight.
  • the first storage unit 63 includes:
  • a first searching module configured to obtain, from the preset classification comparison table, a preset value corresponding to each of the photo storage folders;
  • a numerical comparison module configured to compare the classified attribution value of the photo with the obtained preset value, and determine the preset value that is the smallest absolute value of the difference between the classified attribution values of the photo ;
  • a determining module configured to determine the determined preset value and the photo if the absolute value of the difference between the preset value and the classified attribution value of the photo is located in a preset difference interval Classification attribution values match;
  • a storage module configured to store the photo in the photo storage folder corresponding to the determined preset value.
  • the attribute obtaining unit 61 includes:
  • a feature extraction module configured to extract image features of the photo
  • a type output module configured to input the extracted image feature into the trained convolutional neural network model, and output a photo type of the photo
  • the trained convolutional neural network model is obtained according to the following steps:
  • a convolutional neural network model including an input layer, a convolution layer, a fully connected layer, and an output layer;
  • the network connection weight and the threshold between the nodes of the convolutional neural network model are preset to a random value satisfying the preset condition, and the ideal output value of the sample photo is set.
  • Sample photos are randomly selected from the set number of sample photos, input to the input layer, passed through the convolution layer and the fully connected layer, and transmitted to the output layer to obtain the actual output value of the sample photo, complete a round of training, and calculate the actual The difference between the output value and the ideal output value;
  • the network connection weights and thresholds between the nodes of each layer are adjusted according to the specified learning rules, and the convolutional neural network model is trained again until the calculated difference is not greater than a preset threshold. , complete the training, and obtain a trained convolutional neural network model.
  • the photo storage device further includes:
  • a photo arranging unit 71 configured to arrange photos stored in the same photo storage folder according to a timeline stored in the photo storage folder;
  • the cover determining unit 72 is configured to use the latest photo stored in the folder as the cover of the photo storage folder according to the timeline stored in the photo storage folder.
  • the photographing attribute of the photo by acquiring the photographing attribute of the photo, calculating the classification attribution value of the photo according to the photographing attribute of the photograph and the preset attribute weight, and then searching for the classified attribution value in the preset classification comparison table.
  • Corresponding photo storage folder and storing the photo in a photo storage folder corresponding to the classification attribution value, if a photo storage folder corresponding to the classification attribution value is not found, a storage path is automatically generated.
  • the storage path corresponds to a new photo storage folder, and the photo is stored in the new photo storage folder, and the photos can be classified and stored when the shooting is completed, without the user manually storing the photos, saving the user. Time and improve the efficiency of photo storage.
  • FIG. 8 is a schematic diagram of a server according to an embodiment of the present application.
  • the server 8 of this embodiment includes a processor 80, a memory 81, and computer readable instructions 82, such as a photo storage program, stored in the memory 81 and executable on the processor 80.
  • the processor 80 executes the computer readable instructions 82 to implement the steps in the various photo storage method embodiments described above, such as steps 101 through 104 shown in FIG.
  • the processor 80 when executing the computer readable instructions 82, implements the functions of the various modules/units in the various apparatus embodiments described above, such as the functions of the modules 61-64 shown in FIG.
  • the computer readable instructions 82 may be partitioned into one or more modules/units that are stored in the memory 81 and executed by the processor 80, To complete this application.
  • the one or more modules/units may be a series of computer readable instruction instruction segments capable of performing a particular function for describing the execution of the computer readable instructions 82 in the server 8.
  • the server 8 can be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
  • the server may include, but is not limited to, a processor 80, a memory 81. It will be understood by those skilled in the art that FIG. 8 is merely an example of the server 8, does not constitute a limitation of the server 8, may include more or less components than those illustrated, or combine some components, or different components, such as
  • the server may also include an input and output device, a network access device, a bus, and the like.
  • the processor 80 may be a central processing unit (CPU), or may be another general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the memory 81 may be an internal storage unit of the server 8, such as a hard disk or a memory of the server 8.
  • the memory 81 may also be an external storage device of the server 8, such as a plug-in hard disk equipped with the server 8, a smart memory card (SMC), and a Secure Digital (SD) card. Flash card, etc.
  • the memory 81 may also include both an internal storage unit of the server 8 and an external storage device.
  • the memory 81 is used to store the computer readable instructions and other programs and data required by the server.
  • the memory 81 can also be used to temporarily store data that has been output or is about to be output.
  • the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated modules/units if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium.
  • the present application implements all or part of the processes in the foregoing embodiments, and may also be implemented by computer readable instructions, which may be stored in a computer readable storage medium.
  • the computer readable instructions when executed by a processor, may implement the steps of the various method embodiments described above. It should be noted that the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in a jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, computer readable media Does not include electrical carrier signals and telecommunication signals.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Processing Or Creating Images (AREA)

Abstract

一种照片存储方法、存储介质、服务器（8）和装置，包括：获取照片的拍摄属性（S101）；根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值（S102）；查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中（S103）；若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹（S104）。该方法无需用户手动对照片分类存储，可节约用户的时间，提高照片存储的效率。

Description

一种照片存储方法、存储介质、服务器和装置
本申请要求于2018年05月14日提交中国专利局、申请号为CN 201810456993.1、发明名称为“一种照片存储方法、存储介质和服务器”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，尤其涉及一种照片存储方法、存储介质、服务器和装置。
背景技术
随着科技的发展，带有拍照功能的手机、数码相机等移动终端的体积越来越轻薄，拍出的照片质量也越来越好。人们经常会随身携带具有拍照功能的智能终端，以便随时拍摄照片。
然而，目前具有拍照功能的智能终端，一般是按照拍摄时间存储所拍的照片，有的也可以根据照片来源的APP进行分类，但都不能根据照片中的信息进行自动分类和存储。当用户需要从大量照片中筛选出包括相同照片信息的照片时，只能通过人工操作，一一查找识别，从而浪费了大量的人力和时间。
技术问题
现有技术中，当用户需要从大量照片中筛选出包括相同照片信息的照片时，只能通过人工操作，一一查找识别，从而浪费了大量的人力和时间的问题。
技术解决方案
本申请实施例提供了一种照片存储方法、存储介质、服务器和装置，以解决现有技术中，当用户需要从大量照片中筛选出包括相同照片信息的照片时，只能通过人工操作，一一查找识别，从而浪费了大量的人力和时间的问题。
本申请实施例的第一方面提供了一种照片存储方法，包括：
获取照片的拍摄属性；
根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
本申请实施例的第二方面提供了一种服务器，包括存储器以及处理器，所述存储器存储有可在所述处理器上运行的计算机可读指令，所述处理器执行所述计算机可读指令时实现如下步骤：
获取照片的拍摄属性；
根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
本申请实施例的第三方面提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机可读指令，所述计算机可读指令被处理器执行时实现如下步骤：
获取照片的拍摄属性；
根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
本申请实施例的第四方面提供了一种照片存储装置，包括：
属性获取单元，用于获取照片的拍摄属性；
归属值计算单元，用于根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
第一存储单元，用于查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
第二存储单元，用于若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
有益效果
本申请实施例中，通过获取照片的拍摄属性，根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值，然后查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中，若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹，在拍摄完成时即可将照片分类存储，无需用户手动对照片分类存储，节约了用户的时间，并提高了照片存储的效率。
附图说明
图1是本申请实施例提供的照片存储方法的实现流程图；
图2是本申请实施例提供的照片存储方法S102的具体实现流程图；
图3是本申请实施例提供的照片存储方法B2的具体实现流程图；
图4是本申请实施例提供的照片存储方法S103的具体实现流程图；
图5是本申请另一实施例提供的照片存储方法的实现流程图；
图6是本申请实施例提供的照片存储装置的结构框图；
图7是本申请另一实施例提供的照片存储装置的结构框图；
图8是本申请实施例提供的服务器的示意图。
本发明的实施方式
为使得本申请的发明目的、特征、优点能够更加的明显和易懂，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，下面所描述的实施例仅仅是本申请一部分实施例，而非全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其它实施例，都属于本申请保护的范围。
图1示出了本申请实施例提供的照片存储方法的实现流程，该方法流程包括步骤S101至S104。各步骤的具体实现原理如下：
S101：获取照片的拍摄属性。
具体地，所述拍摄属性包括拍摄时间、拍摄地点和照片类型。其中，照片类型包括人物照、食物照、风景照以及二维码。
在本申请实施例中，照片的拍摄属性可以根据使用智能终端拍摄照片的用户自定义标签确定。例如，2018年1月20日10点，在家拍摄一张自拍人脸照，其中拍摄地点可通过用户自定义地点标签，如定义地理位置为家、工作单位或者某景点。用户可能在每一天的同一个时间同一个地点拍摄照片，例如，去上班时在停车场拍摄的二维码图片。
可选地，通过大数据统计智能终端的位置信息来确定拍摄地点，例如，将一天按工作时间段和休息时间段进行分割，例如，早上九点到下午六点为工作时间段，晚上十一点至第二天早上六点，为休息时间段。获取智能终端在预设的工作日中，各个时间段的地理位置，累计同一个时间段在同一个地理位置的总时长，将工作时间段总时长最长的地理位置的地点标记为工作单位，将休息时间段总时长最长的地理位置的地点标记为家。
可选地，将预设的工作日中智能终端的地理位置进行聚类分析。具体地，将预设的工作日中，所有工作时间段的地理位置存入工作时间集合，随机选取第一指定个数的地理位置为第一聚类中心，计算该工作时间集合中各个地理位置与第一聚类中心的距离值，根据计算的距离值与指定的聚类中心，将该工作时间集合中的地理位置进行初始聚类，从完成 初始聚类后的地理位置中再选取第二指定个数的地理位置作为第二聚类中心，将完成初始聚类后的地理位置以第二聚类中心作为中心簇进行聚类，以此类推，直到该工作时间集合中的聚类中心收敛，定义收敛后工作时间集合的聚类中心(地理位置)的地点标签为工作单位。根据上述同样的方法，定义收敛后工作时间集合的聚类中心(地理位置)的地点标签为家。需说明的是，拍摄地点不仅包括工作单位和家，还包括停车场、学校等，对于其它拍摄地点同样可根据上述方法确定。
可选地，对于照片的照片类型，可通过将照片输入至训练好的卷积神经网络模型中确定。具体地，提取所述照片的图像特征，将提取的照片的图像特征输入到已训练的卷积神经网络模型的输入层，在输出层输出该照片的照片类型。所述已训练好的卷积神经网络模型根据如下步骤获取：
A1：获取设定数量的样本照片，所述样本照片预先设有类型标签。具体地，通过获取设定数量的样本照片，该样本照片都预先设有如风景、人物、食物、二维码等标签。
A2：建立包括输入层、卷积层、全连接层和输出层的卷积神经网络模型。
A3：在首次训练时，将所述卷积神经网络模型各层节点之间的网络连接权值与阈值预先设置成满足预设条件的随机值，并设定所述样本照片的理想输出值，从所述设定数量的样本照片中随机选取样本照片，输入至输入层，经过卷积层和全连接层，传送到输出层，获取所述样本照片的实际输出值，完成一轮训练，并计算实际输出值与理想输出值的差值。
A4：根据计算的差值，按照指定的学习规则对各层节点之间的网络连接权值和阈值进行调整，再次对卷积神经网络模型进行训练，直至当计算的差值不大于预设的阈值时，完成训练，获取训练好的卷积神经网络模型。
具体地，建立包括输入层、卷积层、全连接层和输出层的卷积神经网络模型，训练分如下，从样本照片中随机选取样本输入卷积神经网络模型，计算样本照片的输出值，在且仅在第一次训练时，将卷积神经网络模型各层节点之间的网络连接权值、阈值预先设置成小的接近于0的随机值，并设定样本照片的理想输出值，将样本照片从输入层经过卷积层和全连接层，传送到输出层，获取该样本照片的实际输出值，完成一轮训练，计算实际输出值与理想输出值的差值。在本申请实施例中，根据如下公式计算该卷积神经网络的全局差值D：
Figure PCTCN2018097097-appb-000001
其中，D t为第t张样本照片的理想输出值I t与实际输出值R t的差值，n为正整数，且n为进行训练的样本照片的数量总数。按极小化误差的方法调整权矩阵。设置误差阈值，若D大于该阈值，则按照Delta学习规则对各层节点之间的网络连接权值和阈值进行调整，然后再次对卷积神经网络模型进行训练，直至网络全局误差D不大于该阈值为止，结束训 练，将该次训练的权值和阈值保存作为该卷积神经网络的最优模型参数，得到训练好的卷积神经网络模型。其中，Delta学习规则的学习信号规定为：r＝(dj-f(wTjx))f′(wTjx)＝(dj-oj))f′(netj)。
在本申请实施例中，通过将该设定数量的样本照片输入至卷积神经网络模型进行训练，确定该神经网络模型的最优模型参数，从而获得训练好的卷积神经网络模型，通过将拍摄的照片输入至训练好的卷积神经网络模型即可快速获取所述照片的照片类型，进而提高照片分类存储的效率。
S102：根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值。
具体地，每一种拍摄属性对应一种预设的属性权重，例如，拍摄地点对应预设的地点权重，拍摄时间对应预设的时间权重，照片类型对应预设的类型权重。根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类属性值，从而将所述照片的拍摄属性数值化，便于将所述照片分类存储。
作为本申请的一个实施例，如图2所示，上述S102具体包括：
B1：建立三维坐标系。
B2：根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标。具体地，将所述照片的拍摄属性映射为所述三维坐标系中的一点。
B3：计算所述照片的三维坐标对应的点与所述三维坐标系原点的距离值，将所述距离值作为所述照片的分类归属值。
本申请实施例通过将照片映射为三维坐标系中的一个点，将所述照片在所述三维坐标系对应的点至原点的距离值作为所述照片的分类归属值，将所述照片的拍摄属性数值化，以便根据所述分类归属值进行分类存储。
作为本申请的一个实施例，图3示出了本申请实施例提供的照片存储方法步骤B2的具体实现流程，详述如下：
B21：预先设立映射表，所述映射表中包括不同的拍摄时间分别对应的数值、不同的拍摄地点分别对应的数值以及不同的照片类型对应的数值。具体地，拍摄时间按拍摄日期或者拍摄时刻进行划分，在所述映射表中，若所述拍摄时间按拍摄日期进行划分，拍摄日期对应的数值按指定差值逐日递增；若所述拍摄时间按拍摄时刻进行划分，将拍摄时间划分为若干个时间段，所述时间段对应的数值按指定差值逐段递增。若拍摄时间段例如，上午十点对应的数值为5，下午三点对应的数值为8。对于拍摄地点，不同拍摄地点的经纬度不同，在所述映射表中，所述映射表中包括不同经纬度对应的数值。而拍摄类型通过预先确定经过大数据统计分析确定的照片类型，并针对不同照片类型设置不同的数值，如人物照 为10，风景照为20。
B22：从所述映射表中查找所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值。
B23：根据所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值和所述属性权重，确定所述照片在所述三维坐标系中的三维坐标。具体地，将所述照片的拍摄时间对应的数值与预设的时间权重的乘积，作为所述三维坐标系中第一坐标轴上的值，将所述照片的拍摄地点对应的数值与预设的地点权重的乘积作为所述三维坐标系中第二坐标轴上的值，将所述照片的照片类型对应的数值与预设的类型权重的乘积作为所述三维坐标系中第三坐标轴上的值，从而获得所述照片在所述三维坐标系中的三维坐标。
S103：查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中。
在本申请实施例中，所述预设的分类对照表中包括分类归属值与照片存储文件夹对应的预设数值的对应关系。
具体地，图4示出了本申请实施例提供的照片存储方法S103的具体实现流程，详述如下：
C1：从所述预设的分类对照表中获取每个所述照片存储文件夹对应的预设数值。具体地，所述照片存储文件夹对应的预设数值可根据存入第一张照片的分类归属值确定。
C2：将所述照片的分类归属值与获取到的所述预设数值进行逐一比对，确定与所述照片的分类归属值的差值绝对值最小的所述预设数值。
C3：若确定出的所述预设数值与所述照片的分类归属值的差值绝对值位于预设的差值区间，则判定确定出的所述预设数值与所述照片的分类归属值匹配。当然，若所述预设数值与所述照片的分类归属值的差值绝对值不在所述预设的差值区间之内，则判定确定出的所述预设数值与所述照片的分类归属值不匹配。
C4：将所述照片存入确定出的所述预设数值对应的所述照片存储文件夹中。具体地，将所述照片存入与所述照片的分类归属值匹配的预设数值对应的照片存储文件夹中。
S104：若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
具体地，定义所述预设数值与所述照片的分类归属值的差值为E，当E≤η时，将该照片存入与所述预设数值对应的照片存储文件夹中，当E＞η时，即所述预设的分类对照表中不存在与所述照片的分类归属值的差值绝对值位于预设的差值区间的所述预设数值，此时，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，文件夹说到底就是一个文件的存储路径，即通过自动生成一个存储路径，新建一个照片存储文件夹， 并将该照片存入新建的文件夹中。需说明的是，若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径与已有的照片存储文件夹对应的存储路径不相同。进一步地，将所述照片的分类归属值作为所述照片存储文件的预设数值。
可选地，对于人物类型的照片，根据照片中的人脸的面积值对照片进一步进行分类，例如，进一步细分为自拍照。具体地，对人物照进行人脸识别，并计算识别的人脸的面积值，确定该人脸的面积值在该人物照中所占的面积比例，若该面积比例大于预设的面积比例阈值，则进一步将该人物照存放入自拍照文件夹中，以便用户快速对自拍照进行处理。
可选地，对于同一个照片存储文件夹中的照片，可计算照片的相似度，根据照片的相似度建立所述照片存储文件夹的子文件夹，将相似度大于预设相似度阈值的照片存入同一个子文件夹中。例如，计算照片的相似度判断是否为同一人脸照，若是，则在该照片存储文件夹中建立子文件夹，将同一人脸的照片存入该子文件夹中。
本申请实施例中，通过获取照片的拍摄属性，根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值，然后查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中，若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹，在拍摄完成时即可将照片分类存储，无需用户手动对照片分类存储，节约了用户的时间，并提高了照片存储的效率。
进一步地，基于上述图1实施例中所提供的照片存储方法，提出本申请的另一实施例。在本申请实施例中，在图1所示的步骤S101-S104的基础上，如图5所示，所述照片存储方法还包括：
S201：获取照片的拍摄属性。
S202：根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值。
S203：查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中。
S204：若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
本实施例中，步骤S201至步骤S204的具体步骤参见前述实施例步骤S101至步骤S104，在此不再赘述。
S205：将存储在同一个照片存储文件夹中的照片按存入所述照片存储文件夹的时间线排列。
S206：根据存入所述照片存储文件夹的时间线，将最新存入该文件夹的照片作为所述 照片存储文件夹的封面。
具体地，拍摄时间的优先级高于照片类型，照片类型的优先级高于拍摄地点，将同一个照片存储文件夹中的照片按存入的时间线排列，当所述照片存储文件夹新存入照片时，将新存入的照片替换所述照片存储文件夹的封面，成为所述照片存储文件夹新的封面，实现根据拍摄存储照片动态更新照片存储文件夹的封面，用户无需点开文件夹即可快速清楚文件夹内存储的照片类型，提高用户体验。
本申请实施例中，通过获取照片的拍摄属性，根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值，然后查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中，若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹，在拍摄完成时即可将照片分类存储，无需用户手动对照片分类存储，节约了用户的时间，并提高了照片存储的效率，同时，将存储在同一个照片存储文件夹中的照片按存入所述照片存储文件夹的时间线排列，根据存入所述照片存储文件夹的时间线，将最新存入该文件夹的照片作为所述照片存储文件夹的封面，动态更新封面，用户无需点开文件夹即可快速清楚文件夹内存储的照片类型，提高用户体验。
应理解，上述实施例中各步骤的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
对应于上文实施例所述的照片存储方法，图6示出了本申请实施例提供的照片存储装置的结构框图，为了便于说明，仅示出了与本申请实施例相关的部分。
参照图6，该照片存储装置包括：属性获取单元61，归属值计算单元62，第一存储单元63，第二存储单元64，其中：
属性获取单元61，用于获取照片的拍摄属性；
归属值计算单元62，用于根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
第一存储单元63，用于查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
第二存储单元64，用于若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
可选地，所述归属值计算单元62包括：
坐标系建立模块，用于建立三维坐标系；
坐标确定模块，用于根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标；
归属值确定模块，用于计算所述照片的三维坐标对应的点与所述三维坐标系原点的距离值，将所述距离值作为所述照片的分类归属值。
可选地，所述坐标确定模块包括：
预设子模块，用于预先设立映射表，所述映射表中包括不同的拍摄时间分别对应的数值、不同的拍摄地点分别对应的数值以及不同的照片类型对应的数值；
数值查找子模块，用于从所述映射表中查找所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值；
坐标确定子模块，用于根据所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值和所述属性权重，确定所述照片在所述三维坐标系中的三维坐标。
可选地，所述第一存储单元63包括：
第一查找模块，用于从所述预设的分类对照表中获取每个所述照片存储文件夹对应的预设数值；
数值比对模块，用于将所述照片的分类归属值与获取到的所述预设数值进行逐一比对，确定与所述照片的分类归属值的差值绝对值最小的所述预设数值；
判定模块，用于若确定出的所述预设数值与所述照片的分类归属值的差值绝对值位于预设的差值区间，则判定确定出的所述预设数值与所述照片的分类归属值匹配；
存储模块，用于将所述照片存入确定出的所述预设数值对应的所述照片存储文件夹中。
可选地，所述属性获取单元61包括：
特征提取模块，用于提取所述照片的图像特征；
类型输出模块，用于将提取的图像特征输入至已训练好的卷积神经网络模型，输出所述照片的照片类型；
所述已训练好的卷积神经网络模型根据如下步骤获取：
获取设定数量的样本照片，所述样本照片预先设有类型标签；
建立包括输入层、卷积层、全连接层和输出层的卷积神经网络模型；
在首次训练时，将所述卷积神经网络模型各层节点之间的网络连接权值与阈值预先设置成满足预设条件的随机值，并设定所述样本照片的理想输出值，从所述设定数量的样本照片中随机选取样本照片，输入至输入层，经过卷积层和全连接层，传送到输出层，获取所述样本照片的实际输出值，完成一轮训练，并计算实际输出值与理想输出值的差值；
根据计算的差值，按照指定的学习规则对各层节点之间的网络连接权值和阈值进行调 整，再次对卷积神经网络模型进行训练，直至当计算的差值不大于预设的阈值时，完成训练，获取训练好的卷积神经网络模型。
可选地，如图7所示，所述照片存储装置还包括：
照片排列单元71，用于将存储在同一个照片存储文件夹中的照片按存入所述照片存储文件夹的时间线排列；
封面确定单元72，用于根据存入所述照片存储文件夹的时间线，将最新存入该文件夹的照片作为所述照片存储文件夹的封面。
本申请实施例中，通过获取照片的拍摄属性，根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值，然后查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中，若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹，在拍摄完成时即可将照片分类存储，无需用户手动对照片分类存储，节约了用户的时间，并提高了照片存储的效率。
图8是本申请一实施例提供的服务器的示意图。如图8所示，该实施例的服务器8包括：处理器80、存储器81以及存储在所述存储器81中并可在所述处理器80上运行的计算机可读指令82，例如照片存储程序。所述处理器80执行所述计算机可读指令82时实现上述各个照片存储方法实施例中的步骤，例如图1所示的步骤101至104。或者，所述处理器80执行所述计算机可读指令82时实现上述各装置实施例中各模块/单元的功能，例如图6所示模块61至64的功能。
示例性的，所述计算机可读指令82可以被分割成一个或多个模块/单元，所述一个或者多个模块/单元被存储在所述存储器81中，并由所述处理器80执行，以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机可读指令指令段，该指令段用于描述所述计算机可读指令82在所述服务器8中的执行过程。
所述服务器8可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述服务器可包括，但不仅限于，处理器80、存储器81。本领域技术人员可以理解，图8仅仅是服务器8的示例，并不构成对服务器8的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述服务器还可以包括输入输出设备、网络接入设备、总线等。
所述处理器80可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA) 或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
所述存储器81可以是所述服务器8的内部存储单元，例如服务器8的硬盘或内存。所述存储器81也可以是所述服务器8的外部存储设备，例如所述服务器8上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。进一步地，所述存储器81还可以既包括所述服务器8的内部存储单元也包括外部存储设备。所述存储器81用于存储所述计算机可读指令以及所述服务器所需的其他程序和数据。所述存储器81还可以用于暂时地存储已经输出或者将要输出的数据。
在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请实现上述实施例方法中的全部或部分流程，也可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一计算机可读存储介质中，该计算机可读指令在被处理器执行时，可实现上述各个方法实施例的步骤。需要说明的是，所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读介质不包括电载波信号和电信信号。
以上所述实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围，均应包含在本申请的保护范围之内。

Claims (20)

  1. 一种照片存储方法，其特征在于，包括：
    获取照片的拍摄属性；
    根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
    查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
    若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
  2. 根据权利要求1所述的照片存储方法，其特征在于，所述拍摄属性包括拍摄时间、拍摄地点和照片类型，所述根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值，包括：
    建立三维坐标系；
    根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标；
    计算所述照片的三维坐标对应的点与所述三维坐标系原点的距离值，将所述距离值作为所述照片的分类归属值。
  3. 根据权利要求2所述的照片存储方法，其特征在于，所述根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标，包括：
    预先设立映射表，所述映射表中包括不同的拍摄时间分别对应的数值、不同的拍摄地点分别对应的数值以及不同的照片类型对应的数值；
    从所述映射表中查找所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值；
    根据所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值和所述属性权重，确定所述照片在所述三维坐标系中的三维坐标。
  4. 根据权利要求1所述的照片存储方法，其特征在于，所述查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中，包括：
    从所述预设的分类对照表中获取每个所述照片存储文件夹对应的预设数值；
    将所述照片的分类归属值与获取到的所述预设数值进行逐一比对，确定与所述照 片的分类归属值的差值绝对值最小的所述预设数值；
    若确定出的所述预设数值与所述照片的分类归属值的差值绝对值位于预设的差值区间，则判定确定出的所述预设数值与所述照片的分类归属值匹配；
    将所述照片存入确定出的所述预设数值对应的所述照片存储文件夹中。
  5. 根据权利要求1至4任一项所述的照片存储方法，其特征在于，所述照片存储方法还包括：
    将存储在同一个照片存储文件夹中的照片按存入所述照片存储文件夹的时间线排列；
    根据存入所述照片存储文件夹的时间线，将最新存入该文件夹的照片作为所述照片存储文件夹的封面。
  6. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机可读指令，其特征在于，所述计算机可读指令被处理器执行时实现如下步骤：
    获取照片的拍摄属性；
    根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
    查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
    若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
  7. 根据权利要求6所述的计算机可读存储介质，其特征在于，所述拍摄属性包括拍摄时间、拍摄地点和照片类型，所述根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值，包括：
    建立三维坐标系；
    根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标；
    计算所述照片的三维坐标对应的点与所述三维坐标系原点的距离值，将所述距离值作为所述照片的分类归属值。
  8. 根据权利要求7所述的计算机可读存储介质，其特征在于，所述根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标，包括：
    预先设立映射表，所述映射表中包括不同的拍摄时间分别对应的数值、不同的拍摄地点分别对应的数值以及不同的照片类型对应的数值；
    从所述映射表中查找所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值；
    根据所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值和所述属性权重，确定所述照片在所述三维坐标系中的三维坐标。
  9. 根据权利要求6所述的计算机可读存储介质，其特征在于，所述查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中，包括：
    从所述预设的分类对照表中获取每个所述照片存储文件夹对应的预设数值；
    将所述照片的分类归属值与获取到的所述预设数值进行逐一比对，确定与所述照片的分类归属值的差值绝对值最小的所述预设数值；
    若确定出的所述预设数值与所述照片的分类归属值的差值绝对值位于预设的差值区间，则判定确定出的所述预设数值与所述照片的分类归属值匹配；
    将所述照片存入确定出的所述预设数值对应的所述照片存储文件夹中。
  10. 根据权利要求6至9任一项所述的计算机可读存储介质，其特征在于，所述计算机可读指令被处理器执行时还实现如下步骤：
    将存储在同一个照片存储文件夹中的照片按存入所述照片存储文件夹的时间线排列；
    根据存入所述照片存储文件夹的时间线，将最新存入该文件夹的照片作为所述照片存储文件夹的封面。
  11. 一种服务器，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令，其特征在于，所述处理器执行所述计算机可读指令时实现如下步骤：
    获取照片的拍摄属性；
    根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值；
    查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
    若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
  12. 根据权利要求11所述的服务器，其特征在于，所述拍摄属性包括拍摄时间、拍摄地点和照片类型，所述根据所述照片的拍摄属性和预设的属性权重，计算所述照片的分类归属值，包括：
    建立三维坐标系；
    根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标；
    计算所述照片的三维坐标对应的点与所述三维坐标系原点的距离值，将所述距离值作为所述照片的分类归属值。
  13. 根据权利要求12所述的服务器，其特征在于，所述根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标，包括：
    预先设立映射表，所述映射表中包括不同的拍摄时间分别对应的数值、不同的拍摄地点分别对应的数值以及不同的照片类型对应的数值；
    从所述映射表中查找所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值；
    根据所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值和所述属性权重，确定所述照片在所述三维坐标系中的三维坐标。
  14. 根据权利要求11所述的服务器，其特征在于，所述查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中，包括：
    从所述预设的分类对照表中获取每个所述照片存储文件夹对应的预设数值；
    将所述照片的分类归属值与获取到的所述预设数值进行逐一比对，确定与所述照片的分类归属值的差值绝对值最小的所述预设数值；
    若确定出的所述预设数值与所述照片的分类归属值的差值绝对值位于预设的差值区间，则判定确定出的所述预设数值与所述照片的分类归属值匹配；
    将所述照片存入确定出的所述预设数值对应的所述照片存储文件夹中。
  15. 根据权利要求11至14任一项所述的服务器，其特征在于，所述照片存储方法还包括：
    将存储在同一个照片存储文件夹中的照片按存入所述照片存储文件夹的时间线排列；
    根据存入所述照片存储文件夹的时间线，将最新存入该文件夹的照片作为所述照片存储文件夹的封面。
  16. 一种照片存储装置，其特征在于，包括：
    属性获取单元，用于获取照片的拍摄属性；
    归属值计算单元，用于根据所述照片的拍摄属性和预设的属性权重，计算所述照 片的分类归属值；
    第一存储单元，用于查找预设的分类对照表中所述分类归属值对应的照片存储文件夹，并将所述照片存入所述分类归属值对应的照片存储文件夹中；
    第二存储单元，用于若查找不到所述分类归属值对应的照片存储文件夹，自动生成一个存储路径，所述存储路径对应一个新的照片存储文件夹，并将所述照片存入所述新的照片存储文件夹。
  17. 根据权利要求16所述的照片存储装置，其特征在于，所述归属值计算单元包括：
    坐标系建立模块，用于建立三维坐标系；
    坐标确定模块，用于根据所述拍摄属性对应的所述属性权重，将所述照片的拍摄时间、拍摄地点和照片类型映射至所述三维坐标系中，确定所述照片在所述三维坐标系中的三维坐标；
    归属值确定模块，用于计算所述照片的三维坐标对应的点与所述三维坐标系原点的距离值，将所述距离值作为所述照片的分类归属值。
  18. 根据权利要求17所述的照片存储装置，其特征在于，所述坐标确定模块包括：
    预设子模块，用于预先设立映射表，所述映射表中包括不同的拍摄时间分别对应的数值、不同的拍摄地点分别对应的数值以及不同的照片类型对应的数值；
    数值查找子模块，用于从所述映射表中查找所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值；
    坐标确定子模块，用于根据所述照片的拍摄时间、拍摄地点和照片类型所分别对应的数值和所述属性权重，确定所述照片在所述三维坐标系中的三维坐标。
  19. 根据权利要求16所述的照片存储装置，其特征在于，所述第一存储单元包括：
    第一查找模块，用于从所述预设的分类对照表中获取每个所述照片存储文件夹对应的预设数值；
    数值比对模块，用于将所述照片的分类归属值与获取到的所述预设数值进行逐一比对，确定与所述照片的分类归属值的差值绝对值最小的所述预设数值；
    判定模块，用于若确定出的所述预设数值与所述照片的分类归属值的差值绝对值位于预设的差值区间，则判定确定出的所述预设数值与所述照片的分类归属值匹配；
    存储模块，用于将所述照片存入确定出的所述预设数值对应的所述照片存储文件夹中。
  20. 根据权利要求16至19任一项所述的照片存储装置，其特征在于，所述照片存储装置还包括：
    照片排列单元，用于将存储在同一个照片存储文件夹中的照片按存入所述照片存储文件夹的时间线排列；
    封面确定单元，用于根据存入所述照片存储文件夹的时间线，将最新存入该文件夹的照片作为所述照片存储文件夹的封面。
PCT/CN2018/097097 2018-05-14 2018-07-25 一种照片存储方法、存储介质、服务器和装置 Ceased WO2019218459A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810456993.1 2018-05-14
CN201810456993.1A CN108897757B (zh) 2018-05-14 2018-05-14 一种照片存储方法、存储介质和服务器

Publications (1)

Publication Number Publication Date
WO2019218459A1 true WO2019218459A1 (zh) 2019-11-21

Family

ID=64342970

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/097097 Ceased WO2019218459A1 (zh) 2018-05-14 2018-07-25 一种照片存储方法、存储介质、服务器和装置

Country Status (2)

Country Link
CN (1) CN108897757B (zh)
WO (1) WO2019218459A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114840620A (zh) * 2022-04-01 2022-08-02 广东省国土资源测绘院 举证材料管理方法、智能终端以及存储介质
CN120010773A (zh) * 2025-01-17 2025-05-16 中移动信息技术有限公司 数据开发方法、装置、设备、存储介质及产品

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109508398B (zh) * 2018-11-30 2020-09-15 维沃移动通信有限公司 一种照片分类方法及其终端设备
CN109766461A (zh) * 2018-12-15 2019-05-17 深圳壹账通智能科技有限公司 基于微表情的照片管理方法、装置、计算机设备及介质
CN111414966B (zh) * 2020-03-25 2023-08-15 抖音视界有限公司 分类方法、装置、电子设备及计算机存储介质
CN112395037B (zh) * 2020-12-07 2024-01-05 深圳云天励飞技术股份有限公司 一种动态封面的选取方法、装置、电子设备及存储介质
CN112766139A (zh) * 2021-01-14 2021-05-07 北京迈格威科技有限公司 目标识别方法及装置、存储介质及电子设备
CN115035785A (zh) * 2022-06-17 2022-09-09 云知声智能科技股份有限公司 一种显示照片的方法、装置、电子设备和存储介质
CN118870211A (zh) * 2024-07-29 2024-10-29 深圳市龙之源科技股份有限公司 拍摄内容存储方法、设备及计算机可读存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510205A (zh) * 2009-03-16 2009-08-19 广州市有福数码科技有限公司 实现相片自动聚类的方法、装置及系统
WO2012073421A1 (ja) * 2010-11-29 2012-06-07 パナソニック株式会社 画像分類装置、画像分類方法、プログラム、記録媒体、集積回路、モデル作成装置
CN104598932A (zh) * 2015-02-13 2015-05-06 成都品果科技有限公司 一种基于事件信息的照片分类方法及系统
CN104866500A (zh) * 2014-02-25 2015-08-26 腾讯科技（深圳）有限公司 图片分类展示方法和装置
CN106155592A (zh) * 2016-07-26 2016-11-23 深圳天珑无线科技有限公司 一种照片处理方法及终端

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7663671B2 (en) * 2005-11-22 2010-02-16 Eastman Kodak Company Location based image classification with map segmentation
KR100845892B1 (ko) * 2006-09-27 2008-07-14 삼성전자주식회사 사진 내의 영상 객체를 지리 객체와 매핑하는 방법 및 그시스템
CN103220376B (zh) * 2013-03-30 2014-07-16 清华大学 利用移动终端的位置数据来定位ip位置的方法
CN103870599A (zh) * 2014-04-02 2014-06-18 联想(北京)有限公司 一种拍摄数据的归集方法、装置和电子设备
CN104699842B (zh) * 2015-03-31 2019-03-26 百度在线网络技术（北京）有限公司 图片展示方法和装置
CN105608425B (zh) * 2015-12-17 2019-02-15 小米科技有限责任公司 对照片进行分类存储的方法及装置
CN105760533B (zh) * 2016-03-08 2019-05-03 Oppo广东移动通信有限公司 一种照片管理方法和装置
CN106101224B (zh) * 2016-06-12 2018-07-17 腾讯科技（深圳）有限公司 识别用户所在地理位置的类别的方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510205A (zh) * 2009-03-16 2009-08-19 广州市有福数码科技有限公司 实现相片自动聚类的方法、装置及系统
WO2012073421A1 (ja) * 2010-11-29 2012-06-07 パナソニック株式会社 画像分類装置、画像分類方法、プログラム、記録媒体、集積回路、モデル作成装置
CN104866500A (zh) * 2014-02-25 2015-08-26 腾讯科技（深圳）有限公司 图片分类展示方法和装置
CN104598932A (zh) * 2015-02-13 2015-05-06 成都品果科技有限公司 一种基于事件信息的照片分类方法及系统
CN106155592A (zh) * 2016-07-26 2016-11-23 深圳天珑无线科技有限公司 一种照片处理方法及终端

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114840620A (zh) * 2022-04-01 2022-08-02 广东省国土资源测绘院 举证材料管理方法、智能终端以及存储介质
CN120010773A (zh) * 2025-01-17 2025-05-16 中移动信息技术有限公司 数据开发方法、装置、设备、存储介质及产品

Also Published As

Publication number Publication date
CN108897757A (zh) 2018-11-27
CN108897757B (zh) 2023-08-22

Similar Documents

Publication Publication Date Title
WO2019218459A1 (zh) 一种照片存储方法、存储介质、服务器和装置
US12001475B2 (en) Mobile image search system
US12197543B2 (en) Ephemeral content management
CN105138963A (zh) 图片场景判定方法、装置以及服务器
CN103631819B (zh) 一种图片命名的方法及系统
US9349077B2 (en) Computer-implemented method, a computer program product and a computer system for image processing
CN111368893A (zh) 图像识别方法、装置、电子设备及存储介质
WO2019047672A1 (zh) 信息推荐方法、计算机设备及存储介质
CN104520848B (zh) 按照出席者搜索事件
CN104915351A (zh) 图片排序方法及终端
US20240037142A1 (en) Systems and methods for filtering of computer vision generated tags using natural language processing
US20220391633A1 (en) Automatically generating an image dataset based on object instance similarity
WO2017067485A1 (zh) 一种图片管理方法、装置及一种终端
CN104750688A (zh) 照片管理方法及系统
WO2020211003A1 (zh) 图像处理方法、计算机可读存储介质和计算机设备
CN105159976A (zh) 图像文件的处理方法和系统
CN116662881A (zh) 一种数据自动标注方法及系统
CN112069335A (zh) 图像分类方法、装置、电子设备及存储介质
WO2018028253A1 (zh) 一种信息处理方法、装置及存储介质
CN112069342A (zh) 图像分类方法、装置、电子设备及存储介质
WO2019109246A1 (zh) 图片查找方法和装置
US20250217411A1 (en) Method and electronic device for image search
TWI477995B (zh) 相片分類系統及方法
CN104598932A (zh) 一种基于事件信息的照片分类方法及系统
CN113269141B (zh) 一种图像处理方法及装置

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: 18919337

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: 18919337

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 01.03.2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18919337

Country of ref document: EP

Kind code of ref document: A1