WO2017107431A1 - 一种图片分类方法、装置及电子设备 - Google Patents

一种图片分类方法、装置及电子设备 Download PDF

Info

Publication number
WO2017107431A1
WO2017107431A1 PCT/CN2016/088228 CN2016088228W WO2017107431A1 WO 2017107431 A1 WO2017107431 A1 WO 2017107431A1 CN 2016088228 W CN2016088228 W CN 2016088228W WO 2017107431 A1 WO2017107431 A1 WO 2017107431A1
Authority
WO
WIPO (PCT)
Prior art keywords
picture
picture element
target
historical
information
Prior art date
Application number
PCT/CN2016/088228
Other languages
English (en)
French (fr)
Inventor
周琦
Original Assignee
北京金山安全软件有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京金山安全软件有限公司 filed Critical 北京金山安全软件有限公司
Publication of WO2017107431A1 publication Critical patent/WO2017107431A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10File systems; File servers
    • G06F16/16File or folder operations, e.g. details of user interfaces specifically adapted to file systems
    • G06F16/168Details of user interfaces specifically adapted to file systems, e.g. browsing and visualisation, 2d or 3d GUIs
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content

Definitions

  • the present invention relates to the field of computer technologies, and in particular, to a picture classification method, apparatus, and electronic device.
  • the embodiment of the invention provides a picture classification method, device and electronic device, which solves the problem of how to classify a picture, thereby ensuring the management effect on the picture and improving the efficiency of picture browsing.
  • a first aspect of the embodiments of the present invention provides a picture classification method, which may include:
  • the obtaining, by the user terminal, the target image to be classified and the image information associated with the target image includes:
  • the classification results should include:
  • the picture information includes area information, time information, picture format information, and the target picture that is simplified and processed;
  • the preset picture recognition manner includes at least one of the following identification methods:
  • An object picture element of the target picture is identified based on the target picture of the simplification process.
  • the filtering process is performed on the classification result, and the classification text corresponding to the target image is generated after the screening process, including:
  • a second aspect of the embodiments of the present invention provides a picture classification apparatus, which may include:
  • An information acquiring unit configured to acquire a target image to be classified and a picture information associated with the target image transmitted by the user terminal;
  • a result obtaining unit configured to acquire a picture element in the target picture based on the picture information, and obtain a classification result corresponding to the picture element
  • a text generating unit configured to perform filtering processing on the classification result, and generate a classified text corresponding to the target image after the filtering process
  • a text sending unit configured to send the classified text to the user terminal, so that the user terminal identifies the classified text to perform classified output on the target image.
  • the information acquiring unit is specifically configured to acquire, by using a multiplex transmission manner, a target image to be classified and a picture information associated with the target image that are transmitted by the user terminal.
  • the result obtaining unit includes:
  • a first element acquiring sub-unit configured to adopt a preset picture recognition manner, and acquire a picture element in the target picture based on the picture information; and acquire, in the pre-trained historical picture element set, the picture element At least one historical picture element;
  • a similarity calculation subunit configured to separately calculate an element similarity between each historical picture element and the picture element in the at least one historical picture element
  • a result generating subunit configured to generate, according to each historical picture element in the at least one historical picture element and an element similarity between each historical picture element and the picture element in the at least one historical picture element, to generate the picture element corresponding Classification results.
  • the picture information includes area information, time information, picture format information, and the target picture that is simplified and processed;
  • the preset picture recognition manner includes at least one of the following identification methods:
  • An object picture element of the target picture is identified based on the target picture of the simplification process.
  • the text generating unit includes:
  • a second element acquiring subunit configured to acquire a first historical picture element in the at least one historical picture element, wherein an element similarity between the first historical picture element and the picture element is greater than a preset similarity threshold;
  • a text generation sub-unit configured to determine a classification category corresponding to the first historical picture element, and generate classification text corresponding to the target picture based on the classification category.
  • a third aspect of the embodiments of the present invention provides an electronic device, which may include: a housing, a processor, a memory, a circuit board, an input device, and an output device, wherein the circuit board is disposed inside the space enclosed by the housing, and the processor
  • the memory, the input device, and the output device are disposed on the circuit board;
  • the memory is configured to store the executable program code;
  • the processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory for use in Perform the following steps:
  • the control output device sends the classified text to the user terminal, so that the user terminal performs classification and output on the picture according to the classified text.
  • the processor is configured to: acquire, by using a multiplexing manner, a target image to be classified and a picture information associated with the target image that are transmitted by the user terminal.
  • the processor is configured to acquire a picture element in the target picture based on the picture information, and obtain a classification result corresponding to the picture element:
  • the picture information includes area information, time information, picture format information, and the target picture that is simplified and processed;
  • the preset picture recognition manner includes at least one of the following identification methods:
  • An object picture element of the target picture is identified based on the target picture of the simplification process.
  • the processor is configured to perform screening processing on the classification result by using the following steps, and generate a classification text corresponding to the target image after the filtering process:
  • a fourth aspect of the embodiments of the present invention discloses a storage medium, where the storage medium is used to store an application, and the application is used to execute a picture classification method disclosed in the first aspect of the embodiments of the present invention at runtime. .
  • An embodiment of the fifth aspect of the present invention discloses an application, wherein the application is used to execute the program at runtime A picture classification method disclosed in the first aspect of the invention.
  • the image element in the target image may be acquired based on the image information, and the classification result corresponding to the image element is obtained, and The classification result is subjected to a filtering process to generate a classification text corresponding to the target picture, and finally the classification text is sent to the user terminal, so that the user terminal classifies and outputs the target picture according to the classified text.
  • FIG. 1 is a schematic flowchart of a picture classification method according to an embodiment of the present invention.
  • FIG. 2 is a schematic flowchart of another picture classification method according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of a picture classification apparatus according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of a result obtaining unit according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a text generating unit according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
  • the image classification method provided by the embodiment of the present invention may be applied to a scenario in which a local image in a user terminal is classified, for example, the image classification device acquires a target image to be classified and a picture associated with the target image transmitted by the user terminal. Information, the picture classification device acquires a picture element in the target picture based on the picture information, and acquires a classification result corresponding to the picture element, and the picture classification device performs screening processing on the classification result, and Generating, by the screening process, the classified text corresponding to the target image, the image classification device sending the classified text to the user terminal, so that the user terminal performs classification and output on the target image according to the classified text. Scenes, etc.
  • the process of classifying the picture is realized, and the management effect on the picture is guaranteed, and the picture can be classified according to the classification. Quickly locate the pictures in the user terminal, improving the efficiency of picture browsing.
  • the image classification device may be a background service device that performs image classification processing connected to a user terminal
  • the user terminal may include: a computer, a tablet computer, a smart phone, a notebook computer, a palmtop computer, and a mobile internet device.
  • Terminal equipment such as (MID) that has functions such as local picture storage and transmission.
  • FIG. 1 is a schematic flowchart diagram of a method for classifying pictures according to an embodiment of the present invention. As shown in FIG. 1, the method of the embodiment of the present invention may include the following steps S101 to S104.
  • the image classification device may acquire the target image to be classified and the image information associated with the target image transmitted by the user terminal, where the target image may be selected by the user in the local image stored by the user terminal.
  • the target picture that needs to be classified may also be automatically uploaded to the updated picture of the picture classifying device when the user terminal detects that there is an update of the local picture set in the local storage, and the local picture set may include a photograph taken by a camera of the user terminal, a picture acquired from another user terminal, or a picture obtained from the Internet, and the like
  • the picture information may include area information and time information when the user terminal acquires the target picture, and
  • the image format information of the target image and the target image of the simplification processing, etc., the geographic information may be obtained based on a Global Positioning System (GPS) or a network component carried by the user terminal, and the time is obtained.
  • Information, image format information, and simplification The target image may be acquired based on image properties of the target image by the user terminal.
  • the picture classification device may acquire a picture element in the target picture based on the picture information, where the picture element may include a picture element of a region, a time, a climate, a format, an object, and the like of the target picture,
  • the image classification device may acquire the classification result corresponding to the picture element.
  • the picture classification device may acquire at least one historical picture element associated with the picture element in the trained historical picture element set, which is understandable.
  • the image classification device may perform training on a plurality of historical pictures in advance, and generate a historical picture element set after training, where the historical picture is a plurality of pictures collected by a developer of the picture classification application, and the history is Picture Training can improve the accuracy of picture recognition.
  • the picture classification device calculates an element similarity between each historical picture element and the picture element in the at least one historical picture element, and the picture classification device may be based on each historical picture element in the at least one historical picture element and Generating similarity between each historical picture element and the picture element in the at least one historical picture element, and generating a classification result corresponding to the picture element.
  • S103 Perform a screening process on the classification result, and generate a classification text corresponding to the target image after the screening process;
  • the image classification device may perform a filtering process on the classification result, and generate a classification text corresponding to the target image after the filtering process, where the classified text is used to describe the target image that needs to be classified.
  • Specific classification categories such as: landscapes, people, climatics, etc., and the classified texts can be recognized by the user terminal to further classify the target images.
  • the image classification device may send the classified text to the user terminal, and when the user terminal obtains the classified text, the user terminal may classify and output the target image according to the classified text, and further The user terminal may store the target picture in a classified folder indicated by the classified text. It can be understood that one picture can be stored in multiple classified folders, and when detecting the user When the classification folder of the target image is opened, the user terminal may output and display the target image under the classification folder.
  • the image element in the target image may be acquired based on the image information, and the classification result corresponding to the image element is obtained, and The classification result is subjected to a filtering process to generate a classification text corresponding to the target picture, and finally the classification text is sent to the user terminal, so that the user terminal classifies and outputs the target picture according to the classified text.
  • FIG. 2 is a schematic flowchart diagram of another method for classifying pictures according to an embodiment of the present invention. As shown in FIG. 2, the method of the embodiment of the present invention may include the following steps S201 to S208.
  • the image classification device may acquire the target image to be classified and the image information associated with the target image transmitted by the user terminal, where the target image may be selected by the user in the local image stored by the user terminal.
  • the picture to be classified may also be automatically uploaded to the updated picture of the picture classifying device when the user terminal detects that there is an update of the local picture set in the local storage, and the local picture set may include the user based on the user.
  • the picture information may include area information, time information when the user terminal acquires the target picture, and the The picture format information of the target picture and the target picture of the simplification process, etc., the area information may be obtained based on GPS or network organization carried by the user terminal, the time information, the picture format information, and the simplification processing
  • the target picture of the target picture may be acquired by the user terminal based on a picture attribute of the target picture.
  • the user terminal may transmit, according to the picture classification application, the selected target picture that needs to be classified and the acquired picture information associated with the target picture to the picture classification device, and the transmission manner may be
  • the picture classification device may acquire the target picture to be classified and the picture information associated with the target picture transmitted by the user terminal by using a multiplexing manner.
  • the transmitted target picture and the picture information may be encrypted by the user terminal by using a key agreed in advance with the picture classification device, and the picture classification device may use the key to perform Decryption processing to acquire the target picture and the picture information.
  • the picture classification device may use the key to perform Decryption processing to acquire the target picture and the picture information.
  • the picture element may include a picture element of a region, a time, a climate, a format, an object, and the like of the target picture
  • the picture classification device may adopt a preset picture recognition manner, and acquire the target based on the picture information.
  • the preset picture identification manner may include at least one of: identifying a regional picture element of the target picture based on the area information; and identifying a time picture element of the target picture based on the time information Identifying a climate picture element of the target picture based on the area information and the time information; identifying a format picture element of the target picture based on the picture format information; identifying based on the target picture of the simplified process An object picture element of the target picture.
  • the picture classification device may acquire at least one historical picture element associated with the picture element in a pre-trained set of historical picture elements. It may be understood that the picture classification device may pre-pair multiple historical pictures.
  • the slice is trained, and after the training, a historical picture element set is generated.
  • the historical picture is a plurality of pictures collected by a developer of the picture classification application, and the training of the historical picture can improve the accuracy of the picture recognition.
  • the image classification device calculates an element similarity between each historical picture element and the picture element in the at least one historical picture element, for example, for an object picture element of the target picture, it can be understood that the object picture element Is determined based on the target image of the simplification processing, and thus is an edged shape of the object, the image classification device may acquire at least one historical picture element similar to the object picture element, and separately calculate the image with the object The elemental similarity of the shape of the element.
  • the picture classification device may generate a classification result corresponding to the picture element based on the historical picture elements and element similarities between the historical picture elements and the picture elements.
  • S206 Acquire a first historical picture element in the at least one historical picture element, where an element similarity between the first historical picture element and the picture element is greater than a preset similarity threshold;
  • S207 Determine a classification category corresponding to the first historical picture element, and generate a classification text corresponding to the target picture based on the classification category.
  • the picture classification device may be configured with a similarity threshold in advance, and the picture classification device acquires, in the at least one historical picture element, a first historical picture element whose element similarity is greater than a preset similarity threshold, the picture
  • the classification device may determine a classification category corresponding to the first historical picture element, and generate classification text corresponding to the target picture based on the classification category.
  • the classified text is used to describe a specific classification category for the target picture that needs to be classified, for example, a landscape class, a character class, a climatology class, etc., and the classified text can be recognized by the user terminal, The picture is further classified.
  • the picture classification device may send the classified text to the user terminal.
  • the picture classification device may perform encryption processing on the classified text by using the agreed key, and adopt multiple The road transmission mode is sent to the user terminal, and when the user terminal obtains the classified text, the user terminal may perform decryption processing on the classified text, and classify and output the target image according to the classified text.
  • the The user terminal may store the target picture in the classified folder indicated by the classified text. It can be understood that one picture can be stored in multiple classified folders, when detecting that the user is located on the target picture. When the classification folder is opened, the user terminal may output and display the target picture under the classification folder.
  • the image element in the target image may be acquired based on the image information, and the classification result corresponding to the image element is obtained, and The classification result is subjected to a filtering process to generate a classification text corresponding to the target picture, and finally the classification text is sent to the user terminal, so that the user terminal classifies and outputs the target picture according to the classified text.
  • the picture is classified. The process ensures the management effect on the picture.
  • the picture in the user terminal can be quickly located according to the classification, which improves the efficiency of picture browsing; the picture, picture information and classification are adopted by using multiple channels.
  • the text is transmitted, which can avoid the influence of the network environment during the transmission process, ensure the stability of the picture, picture information and classified text transmission, and can improve the transmission efficiency of the picture, picture information and classified text; by adopting the advance in the transmission process
  • the agreed key encrypts and decrypts the picture, picture information and classified text to ensure the security of the picture, picture information and classified text in the transmission process; classifying the picture by adding the regional information, enriching the category for classifying the picture It improves the accuracy of classifying pictures; by training historical pictures, the accuracy of picture recognition can be improved.
  • FIG. 3 to FIG. 5 The picture sorting apparatus shown in FIG. 3 to FIG. 5 is used to execute the method of the embodiment shown in FIG. 1 and FIG. 2 of the present invention.
  • FIG. 1 and FIG. 2 of the present invention For the convenience of description, only the embodiment of the present invention is shown. For the part, the specific technical details are not disclosed, please refer to the embodiment shown in FIG. 1 and FIG. 2 of the present invention.
  • FIG. 3 is a schematic structural diagram of a picture classification apparatus according to an embodiment of the present invention.
  • the picture classification apparatus 1 of the embodiment of the present invention may include: an information acquisition unit 11, a result acquisition unit 12, a text generation unit 13, and a text transmission unit 14.
  • the information acquiring unit 11 is configured to acquire a target image to be classified and a picture information associated with the target image transmitted by the user terminal;
  • the information acquiring unit 11 may acquire a target image to be classified and a picture information associated with the target picture, where the target picture may be a local picture stored by the user at the user terminal.
  • the selected picture that needs to be classified may also be automatically uploaded to the updated picture of the picture classifying device 1 when the user terminal detects that there is an update of the local picture set in the local storage, and the local picture set may be And including the photograph taken by the camera of the user terminal, the image acquired from another user terminal, or the image obtained from the Internet, and the like
  • the image information may include the regional information when the user terminal acquires the target image,
  • the area information may be obtained based on GPS or network organization carried by the user terminal, the time information, the picture Format information and the target picture of the simplification processing may be by the user Properties based on the image side of the target picture is acquired.
  • the user terminal may transmit, according to the picture classification application, the selected target picture that needs to be classified and the acquired picture information associated with the target picture to the picture classification device 1, and the manner of transmission
  • the multiplex mode may be adopted, and the information acquiring unit 11 may acquire the target image to be classified and the picture information associated with the target picture transmitted by the user terminal by using a multiplex transmission manner.
  • the transmission of the target picture and the picture information can avoid the influence of the network environment during the transmission process, ensure the stability of the picture and picture information transmission, and improve the transmission efficiency of the picture and picture information.
  • the transmitted target picture and the picture information may be encrypted by the user terminal by using a key agreed with the picture classification device 1 in advance, and the information acquiring unit 11 may adopt the secret.
  • the key performs a decryption process to acquire the target picture and the picture information.
  • a result obtaining unit 12 configured to acquire a picture element in the target picture based on the picture information, and obtain a classification result corresponding to the picture element;
  • the result obtaining unit 12 may acquire a picture element in the target picture based on the picture information, where the picture element may include a picture element of a region, a time, a climate, a format, an object, and the like of the target picture.
  • the result obtaining unit 12 may obtain the classification result corresponding to the picture element.
  • the result obtaining unit 12 may acquire at least one historical picture element associated with the picture element in the trained historical picture element set. It can be understood that the picture classification device 1 can train a plurality of historical pictures in advance, and generate a historical picture element set after training, where the historical picture is collected by a developer of the picture classification application. Pictures, by training historical pictures, can improve the accuracy of picture recognition.
  • the result obtaining unit 12 calculates an element similarity between each historical picture element and the picture element in the at least one historical picture element, and the result obtaining unit 12 may be based on each historical picture in the at least one historical picture element. An element and an element similarity between each historical picture element and the picture element in the at least one historical picture element, and a classification result corresponding to the picture element is generated.
  • FIG. 4 is a schematic structural diagram of a result obtaining unit according to an embodiment of the present invention.
  • the result obtaining unit 12 may include:
  • the first element acquisition sub-unit 121 is configured to adopt a preset picture recognition manner, and acquire a picture element in the target picture based on the picture information;
  • the picture element may include a picture element of a region, a time, a climate, a format, an object, and the like of the target picture, and the first element acquisition sub-unit 121 may adopt a preset picture recognition mode, and based on the picture.
  • the preset picture identification manner may include at least one of: identifying a regional picture element of the target picture based on the area information; and identifying the target based on the time information a temporal picture element of the picture; identifying a climate picture element of the target picture based on the area information and the time information; identifying a format picture element of the target picture based on the picture format information; based on the simplified processing
  • the target picture identifies an object picture element of the target picture.
  • the first element acquisition sub-unit 121 is further configured to acquire the map with the pre-trained historical picture element set. At least one historical picture element associated with the slice element;
  • the similarity calculation sub-unit 122 is configured to separately calculate an element similarity between each historical picture element and the picture element in the at least one historical picture element;
  • a result generating sub-unit 123 configured to generate the picture element based on each historical picture element in the at least one historical picture element and an element similarity between each historical picture element and the picture element in the at least one historical picture element Corresponding classification result;
  • the first element acquiring sub-unit 121 may acquire at least one historical picture element associated with the picture element in a pre-trained historical picture element set. It may be understood that the picture classifying apparatus 1 may The plurality of historical pictures are trained in advance, and the historical picture element set is generated after the training, and the historical picture is a plurality of pictures collected by the developer corresponding to the picture classification, and the picture recognition can be improved by training the historical picture. The accuracy.
  • the similarity calculation sub-unit 122 calculates an element similarity between each historical picture element and the picture element in the at least one historical picture element, for example, for an object picture element of the target picture, it can be understood that The object picture element is identified based on the target picture of the simplification process, and thus it is an edged shape of the object, and the first element acquisition sub-unit 121 can acquire at least one historical picture element similar to the object picture element, The similarity calculation sub-unit 122 calculates the element similarity to the shape of the object picture element, respectively.
  • the result generation sub-unit 123 may generate a classification result corresponding to the picture element based on the historical picture elements and element similarities between the historical picture elements and the picture elements.
  • a text generating unit 13 is configured to perform filtering processing on the classification result, and generate a classified text corresponding to the target image after the filtering process;
  • the text generating unit 13 may perform a filtering process on the classification result, and generate a classification text corresponding to the target image after the filtering process, where the classified text is used to describe the target that needs to be classified.
  • a specific classification category of the picture for example, a landscape class, a character class, a climatic class, etc., and the classified text can be recognized by the user terminal to further classify the target picture.
  • FIG. 5 is a schematic structural diagram of a text generating unit according to an embodiment of the present invention.
  • the text generating unit 13 may include:
  • a second element acquiring sub-unit 131 configured to acquire, in the at least one historical picture element, a first historical picture element, where an element similarity between the first historical picture element and the picture element is greater than a preset similarity threshold;
  • a text generation sub-unit 132 configured to determine a classification category corresponding to the first historical picture element, and generate a classification text corresponding to the target picture based on the classification category;
  • the picture classification device 1 may be preset with a similarity threshold, and the second element acquisition sub-unit 131 acquires a first history in which the element similarity is greater than a preset similarity threshold in the at least one historical picture element.
  • the text generation sub-unit 132 may determine a classification category corresponding to the first historical picture element, and generate classification text corresponding to the target picture based on the classification category.
  • the classified text is used to describe a specific classification category for the target picture that needs to be classified, for example, a landscape class, a character class, a climatology class, etc., and the classified text can be recognized by the user terminal, The target picture is further classified.
  • a text sending unit 14 configured to send the classified text to the user terminal
  • the text sending unit 14 may send the classified text to the user terminal.
  • the text sending unit 14 may perform encryption processing on the classified text by using the key of the foregoing agreement. And transmitting to the user terminal by using a multiplexing manner, when the user terminal obtains the classified text, the user terminal may perform decryption processing on the classified text, and classify and output the target image according to the classified text, further
  • the user terminal may store the target picture in a classified folder indicated by the classified text. It can be understood that one picture can be stored in multiple classified folders, when detecting the user to the target When the classification folder of the picture is opened, the user terminal may output and display the target picture under the category folder.
  • the image element in the target image may be acquired based on the image information, and the classification result corresponding to the image element is obtained, and The classification result is subjected to a filtering process to generate a classification text corresponding to the target picture, and finally the classification text is sent to the user terminal, so that the user terminal classifies and outputs the target picture according to the classified text.
  • FIG. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
  • the electronic device of the embodiment of the present invention may include: a housing 600, a circuit board 601, a processor 602, a communication bus 603, an input device 604, an output device 605, and a memory 606, wherein the circuit board 601 Disposed within the space enclosed by the housing 600, a processor 602, a communication bus 603, a memory 606, an input device 604, and an output device 605 are disposed on the circuit board 601; the communication bus 603 is used to implement connection communication between these components.
  • the memory 606 can be a high speed RAM
  • the memory may also be a non-volatile memory such as at least one disk storage.
  • the memory 606 can optionally also be at least one storage device located remotely from the aforementioned processor 602.
  • the processor 602 can be combined with the picture classification device described in FIG. 3 to FIG. 5, and the executable program code is stored in the memory 606; and the processor 602 runs and executes the program by reading the executable program code stored in the memory 606.
  • the program corresponding to the code is used to perform the following steps:
  • the control input device 604 acquires a target picture to be classified and a picture information associated with the target picture transmitted by the user terminal;
  • the control output device 605 transmits the classified text to the user terminal such that the user terminal recognizes the classified text to classify and output the target picture.
  • the processor 602 calls the executable program code stored in the memory 606 to execute the control input device 604 to acquire the target picture to be classified and the picture information associated with the target picture transmitted by the user terminal. To do the following:
  • the control input device 604 acquires a target picture to be classified and a picture information associated with the target picture transmitted by the user terminal in a multiplexing manner.
  • the processor 602 calls the executable program code stored in the memory 606 to perform acquiring the picture element in the target picture based on the picture information, and acquiring the classification result corresponding to the picture element. To do the following:
  • the picture information includes area information, time information, picture format information, and the target picture that is simplified and processed;
  • the preset picture recognition manner includes at least one of the following identification methods:
  • An object picture element of the target picture is identified based on the target picture of the simplification process.
  • processor 602 calls the executable program code stored in the memory 606 to perform a filtering process on the classification result to generate the classified text corresponding to the target image, the following operations are specifically performed:
  • the image element in the target image may be acquired based on the image information, and the classification result corresponding to the image element is obtained, and The classification result is subjected to a filtering process to generate a classification text corresponding to the target picture, and finally the classification text is sent to the user terminal, so that the user terminal classifies and outputs the target picture according to the classified text.
  • the embodiment of the present invention further provides a storage medium, wherein the storage medium is used to store an application, and the application is configured to execute the picture classification method provided above at runtime.
  • An embodiment of the present invention further provides an application, where the application is configured to execute the picture classification method provided above at runtime.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Human Computer Interaction (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Image Analysis (AREA)

Abstract

一种图片分类方法、装置(1)及电子设备,其中方法包括如下步骤:获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息(S101,S201);基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果(S102);对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本(S103);将所述分类文本发送至所述用户终端(S104,S208),以使得所述用户终端识别所述分类文本以对所述目标图片进行分类输出。采用所述方法,可以实现对图片进行分类的过程,进而保证对图片的管理效果,提升图片浏览的效率。

Description

一种图片分类方法、装置及电子设备
相关申请的交叉引用
本申请基于申请号为201510977767.4,申请日为2015年12月23日的中国专利申请提出,并要求该中国专利申请的优先权,该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本发明涉及计算机技术领域,尤其涉及一种图片分类方法、装置及电子设备。
背景技术
随着计算机技术不断的开发和完善,现有的手机和平板电脑等用户终端中,可以将通过摄像、文件传输、网页浏览等方式获得的图片保存在本地存储中,便于用户随时对本地图片进行浏览。然而随着本地图片数量的增加,针对当前需要浏览的本地图片,往往需要耗费较长的时间手动在本地图片集合中进行查找,影响了对图片的管理效果,进而影响了图片浏览的效率。
发明内容
本发明实施例提供一种图片分类方法、装置及电子设备,解决了如何实现对图片进行分类的问题,进而保证对图片的管理效果,提升图片浏览的效率。
本发明实施例第一方面提供了一种图片分类方法,可包括:
获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本;
将所述分类文本发送至所述用户终端,以使得所述用户终端识别所述分类文本以对所述目标图片进行分类输出。
其中,所述获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息,包括:
通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息。
其中,所述基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对 应的分类结果,包括:
采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;
在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
其中,所述图片信息包括地域信息、时间信息、图片格式信息、简略化处理的所述目标图片;
所述预设图片识别方式包括以下至少一个识别方式:
基于所述地域信息识别所述目标图片的地域图片元素;
基于所述时间信息识别所述目标图片的时间图片元素;
基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;
基于所述图片格式信息识别所述目标图片的格式图片元素;
基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
其中,所述对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本,包括:
在所述至少一个历史图片元素中获取第一历史图片元素,其中所述第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。
本发明实施例第二方面提供了一种图片分类装置,可包括:
信息获取单元,用于获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
结果获取单元,用于基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
文本生成单元,用于对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本;
文本发送单元,用于将所述分类文本发送至所述用户终端,以使得所述用户终端识别所述分类文本以对所述目标图片进行分类输出。
其中,所述信息获取单元具体用于通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息。
其中,所述结果获取单元包括:
第一元素获取子单元,用于采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;且在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
相似度计算子单元,用于分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
结果生成子单元,用于基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
其中,所述图片信息包括地域信息、时间信息、图片格式信息、简略化处理的所述目标图片;
所述预设图片识别方式包括以下至少一个识别方式:
基于所述地域信息识别所述目标图片的地域图片元素;
基于所述时间信息识别所述目标图片的时间图片元素;
基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;
基于所述图片格式信息识别所述目标图片的格式图片元素;
基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
其中,所述文本生成单元包括:
第二元素获取子单元,用于在所述至少一个历史图片元素中获取第一历史图片元素,其中所述第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
文本生成子单元,用于确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。
本发明实施例第三方面提供了一种电子设备,可包括:壳体、处理器、存储器、电路板、输入设备和输出设备,其中,电路板安置在壳体围成的空间内部,处理器、存储器、输入设备和输出设备设置在电路板上;存储器用于存储可执行程序代码;处理器通过读取存储器中存储的可执行程序代码来运行与可执行程序代码对应的程序,以用于执行以下步骤:
控制输入设备获取用户终端传输的待分类的图片以及与所述图片相关联的图片信息;
基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
对所述分类结果进行筛选处理,以生成所述目标图片对应的分类文本;
控制输出设备将所述分类文本发送至所述用户终端,以使所述用户终端按照所述分类文本对所述图片进行分类输出。
其中,所述处理器用于:通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息。
其中,所述处理器用于通过以下步骤基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果:
采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;
在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
其中,所述图片信息包括地域信息、时间信息、图片格式信息、简略化处理的所述目标图片;
所述预设图片识别方式包括以下至少一个识别方式:
基于所述地域信息识别所述目标图片的地域图片元素;
基于所述时间信息识别所述目标图片的时间图片元素;
基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;
基于所述图片格式信息识别所述目标图片的格式图片元素;
基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
其中,所述处理器用于通过以下步骤对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本:
在所述至少一个历史图片元素中获取第一历史图片元素,其中所述第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。
本发明实施例第四方面公开了一种存储介质,其中,所述存储介质用于存储应用程序,所述应用程序用于在运行时执行本发明实施例第一方面公开的一种图片分类方法。
本发明实施例第五方面公开了一种应用程序,其中,所述应用程序用于在运行时执行本 发明实施例第一方面公开的一种图片分类方法。
在本发明实施例中,通过获取用户终端传输的需要进行分类的目标图片以及该目标图片关联的图片信息,可以基于图片信息获取目标图片中的图片元素,获取图片元素对应的分类结果,并对分类结果进行筛选处理以生成该目标图片对应的分类文本,最终将分类文本发送至用户终端,以使用户终端按照该分类文本对该目标图片进行分类输出。通过获取图片关联的图片信息,并依据图片信息获取的图片元素对图片进行相应分类,实现了对图片进行分类的过程,保证了对图片的管理效果,由于对图片进行了分类处理,可以依据分类快速定位用户终端中的图片,提升了图片浏览的效率。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的一种图片分类方法的流程示意图;
图2是本发明实施例提供的另一种图片分类方法的流程示意图;
图3是本发明实施例提供的一种图片分类装置的结构示意图;
图4是本发明实施例提供的结果获取单元的结构示意图;
图5是本发明实施例提供的文本生成单元的结构示意图;
图6是本发明实施例提供的一种电子设备的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,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他实施例,都属于本发明保护的范围。
本发明实施例提供的图片分类方法可以应用于对用户终端中的本地图片进行分类处理的场景,例如:图片分类装置获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息,所述图片分类装置基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果,所述图片分类装置对所述分类结果进行筛选处理,并在 筛选处理后生成所述目标图片对应的分类文本,所述图片分类装置将所述分类文本发送至所述用户终端,以使所述用户终端按照所述分类文本对所述目标图片进行分类输出的场景等。通过获取图片关联的图片信息,并依据图片信息获取的图片元素对图片进行相应分类,实现了对图片进行分类的过程,保证了对图片的管理效果,由于对图片进行了分类处理,可以依据分类快速定位用户终端中的图片,提升了图片浏览的效率。
本发明实施例涉及的图片分类装置可以为与用户终端相连接的进行图片分类处理的后台服务设备,所述用户终端可以包括:计算机、平板电脑、智能手机、笔记本电脑、掌上电脑以及移动互联网设备(MID)等具备本地图片存储及传输等功能在内的终端设备。
下面将结合附图1和附图2,对本发明实施例提供的图片分类方法进行详细介绍。
请参见图1,为本发明实施例提供了一种图片分类方法的流程示意图。如图1所示,本发明实施例的所述方法可以包括以下步骤S101-步骤S104。
S101,获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
具体的,图片分类装置可以获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息,所述目标图片可以为用户在所述用户终端所存储的本地图片中所选择的需要进行分类的目标图片,也可以为所述用户终端在检测到本地存储中的本地图片集合存在更新时,自动上传至图片分类装置的所更新的图片,所述本地图片集合可以包括基于所述用户终端的摄像头所拍摄的照片、从其它用户终端获取的图片或者从互联网中获取的图片等,所述图片信息可以包括所述用户终端获取到所述目标图片时的地域信息、时间信息,以及所述目标图片的图片格式信息和简略化处理的所述目标图片等,所述地域信息可以基于所述用户终端携带的全球定位系统(Global Positioning System,GPS)或网络组建等获得,所述时间信息、所述图片格式信息以及简略化处理的所述目标图片可以由所述用户终端基于所述目标图片的图片属性进行获取。
S102,基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
具体的,所述图片分类装置可以基于所述图片信息获取所述目标图片中的图片元素,所述图片元素可以包括所述目标图片的地域、时间、气候、格式、物体等图片元素,所述图片分类装置可以获取所述图片元素对应的分类结果,进一步的,所述图片分类装置可以在所训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素,可以理解的是,所述图片分类装置可以预先对多个历史图片进行训练,并在训练后生成历史图片元素集合,所述历史图片为所述图片分类应用的开发人员所收集的多个图片,通过对历史图片进行 训练,可以提升图片识别的准确性。所述图片分类装置分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,所述图片分类装置可以基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
S103,对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本;
具体的,所述图片分类装置可以对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本,所述分类文本用于描述针对需要进行分类的所述目标图片的具体分类类别,例如:风景类、人物类、气候类等,并且所述分类文本可以被所述用户终端所识别,以对所述目标图片进行进一步的分类处理。
S104,将所述分类文本发送至所述用户终端;
具体的,所述图片分类装置可以将所述分类文本发送至所述用户终端,所述用户终端在获取到所述分类文本时,可以按照所述分类文本对所述目标图片进行分类输出,进一步的,所述用户终端可以将所述目标图片存放至所述分类文本所指示的分类文件夹中,可以理解的是,一个图片可以存放于多个分类文件夹中,当检测到用户对所述目标图片所在的分类文件夹的打开操作时,所述用户终端可以在该分类文件夹下对所述目标图片进行输出显示。
在本发明实施例中,通过获取用户终端传输的需要进行分类的目标图片以及该目标图片关联的图片信息,可以基于图片信息获取目标图片中的图片元素,获取图片元素对应的分类结果,并对分类结果进行筛选处理以生成该目标图片对应的分类文本,最终将分类文本发送至用户终端,以使用户终端按照该分类文本对该目标图片进行分类输出。通过获取图片关联的图片信息,并依据图片信息获取的图片元素对图片进行相应分类,实现了对图片进行分类的过程,保证了对图片的管理效果,由于对图片进行了分类处理,可以依据分类快速定位用户终端中的图片,提升了图片浏览的效率。
请参见图2,为本发明实施例提供了另一种图片分类方法的流程示意图。如图2所示,本发明实施例的所述方法可以包括以下步骤S201-步骤S208。
S201,获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
具体的,图片分类装置可以获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息,所述目标图片可以为用户在所述用户终端所存储的本地图片中所选择的需要进行分类的图片,也可以为所述用户终端在检测到本地存储中的本地图片集合存在更新时,自动上传至图片分类装置的所更新的图片,所述本地图片集合可以包括基于所述用户终 端的摄像头所拍摄的照片、从其它用户终端获取的图片或者从互联网中获取的图片等,所述图片信息可以包括所述用户终端获取到所述目标图片时的地域信息、时间信息,以及所述目标图片的图片格式信息和简略化处理的所述目标图片等,所述地域信息可以基于所述用户终端携带的GPS或网络组建等获得,所述时间信息、所述图片格式信息以及简略化处理的所述目标图片可以由所述用户终端基于所述目标图片的图片属性进行获取。
优选的,所述用户终端可以基于图片分类应用将所选择的需要进行分类的所述目标图片以及所获取的与所述目标图片相关联的图片信息传输至所述图片分类装置,传输的方式可以采用多路传输方式,所述图片分类装置可以通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息。通过采用多路传输方式对所述目标图片和所述图片信息进行传输,可以避免传输过程中受到网络环境的影响,保证了图片及图片信息传输的稳定性,同时可以提升图片及图片信息的传输效率。
进一步的,所传输的所述目标图片和所述图片信息可以经所述用户终端采用预先与所述图片分类装置约定的密钥进行加密处理,所述图片分类装置则可以采用所述密钥进行解密处理以获取所述目标图片和所述图片信息。通过在传输过程中采用预先约定的密钥对图片和图片信息进行加解密,可以保证图片和图片信息在传输过程中的安全性。
S202,采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;
具体的,所述图片元素可以包括所述目标图片的地域、时间、气候、格式、物体等图片元素,所述图片分类装置可以采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素,所述预设图片识别方式可以包括以下至少一个识别方式:基于所述地域信息识别所述目标图片的地域图片元素;基于所述时间信息识别所述目标图片的时间图片元素;基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;基于所述图片格式信息识别所述目标图片的格式图片元素;基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
S203,在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
S204,分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
S205,基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果;
具体的,所述图片分类装置可以在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素,可以理解的是,所述图片分类装置可以预先对多个历史图 片进行训练,并在训练后生成历史图片元素集合,所述历史图片为所述图片分类应用的开发人员所收集的多个图片,通过对历史图片进行训练,可以提升图片识别的准确性。所述图片分类装置分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,例如:针对所述目标图片的物体图片元素,可以理解的是,物体图片元素是基于简略化处理的所述目标图片所识别的,因此其为物体的边缘化形状,所述图片分类装置可以获取与该物体图片元素相似的至少一个历史图片元素,并分别计算与该物体图片元素的形状的元素相似度。所述图片分类装置可以基于所述各历史图片元素以及所述各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
S206,在所述至少一个历史图片元素中获取第一历史图片元素,第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
S207,确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本;
具体的,所述图片分类装置可以预先设置有相似度阈值,所述图片分类装置在所述至少一个历史图片元素中获取元素相似度大于预设相似度阈值的第一历史图片元素,所述图片分类装置可以确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。其中,所述分类文本用于描述针对需要进行分类的所述目标图片的具体分类类别,例如:风景类、人物类、气候类等,并且所述分类文本可以被所述用户终端所识别,以对所述图片进行进一步的分类处理。
S208,将所述分类文本发送至所述用户终端;
具体的,所述图片分类装置可以将所述分类文本发送至所述用户终端,优选的,所述图片分类装置可以采用上述约定的所述密钥对所述分类文本进行加密处理,并采用多路传输方式发送至所述用户终端,所述用户终端在获取到所述分类文本时,可以对其进行解密处理,并按照所述分类文本对所述目标图片进行分类输出,进一步的,所述用户终端可以将所述目标图片存放至所述分类文本所指示的分类文件夹中,可以理解的是,一个图片可以存放于多个分类文件夹中,当检测到用户对所述目标图片所在的分类文件夹的打开操作时,所述用户终端可以在该分类文件夹下对所述目标图片进行输出显示。
在本发明实施例中,通过获取用户终端传输的需要进行分类的目标图片以及该目标图片关联的图片信息,可以基于图片信息获取目标图片中的图片元素,获取图片元素对应的分类结果,并对分类结果进行筛选处理以生成该目标图片对应的分类文本,最终将分类文本发送至用户终端,以使用户终端按照该分类文本对该目标图片进行分类输出。通过获取图片关联的图片信息,并依据图片信息获取的图片元素对图片进行相应分类,实现了对图片进行分类 的过程,保证了对图片的管理效果,由于对图片进行了分类处理,可以依据分类快速定位用户终端中的图片,提升了图片浏览的效率;通过采用多路传输方式对图片、图片信息及分类文本进行传输,可以避免传输过程中受到网络环境的影响,保证了图片、图片信息及分类文本传输的稳定性,同时可以提升图片、图片信息及分类文本的传输效率;通过在传输过程中采用预先约定的密钥对图片、图片信息及分类文本进行加解密,可以保证图片、图片信息及分类文本在传输过程中的安全性;通过加入地域信息对图片进行分类,丰富了对图片进行分类的类别,提升了对图片进行分类的准确性;通过对历史图片进行训练,可以提升图片识别的准确性。
下面将结合附图3-附图5,对本发明实施例提供的图片分类装置进行详细介绍。需要说明的是,附图3-附图5所示的图片分类装置,用于执行本发明图1和图2所示实施例的方法,为了便于说明,仅示出了与本发明实施例相关的部分,具体技术细节未揭示的,请参照本发明图1和图2所示的实施例。
请参见图3,为本发明实施例提供了一种图片分类装置的结构示意图。如图3所示,本发明实施例的所述图片分类装置1可以包括:信息获取单元11、结果获取单元12、文本生成单元13和文本发送单元14。
信息获取单元11,用于获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
具体实现中,所述信息获取单元11可以获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息,所述目标图片可以为用户在所述用户终端所存储的本地图片中所选择的需要进行分类的图片,也可以为所述用户终端在检测到本地存储中的本地图片集合存在更新时,自动上传至图片分类装置1的所更新的图片,所述本地图片集合可以包括基于所述用户终端的摄像头所拍摄的照片、从其它用户终端获取的图片或者从互联网中获取的图片等,所述图片信息可以包括所述用户终端获取到所述目标图片时的地域信息、时间信息,以及所述目标图片的图片格式信息和简略化处理的所述目标图片等,所述地域信息可以基于所述用户终端携带的GPS或网络组建等获得,所述时间信息、所述图片格式信息以及简略化处理的所述目标图片可以由所述用户终端基于所述目标图片的图片属性进行获取。
优选的,所述用户终端可以基于图片分类应用将所选择的需要进行分类的所述目标图片以及所获取的与所述目标图片相关联的图片信息传输至所述图片分类装置1,传输的方式可以采用多路传输方式,所述信息获取单元11可以通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息。通过采用多路传输方式对所述 目标图片和所述图片信息进行传输,可以避免传输过程中受到网络环境的影响,保证了图片及图片信息传输的稳定性,同时可以提升图片及图片信息的传输效率。
进一步的,所传输的所述目标图片和所述图片信息可以经所述用户终端采用预先与所述图片分类装置1约定的密钥进行加密处理,所述信息获取单元11则可以采用所述密钥进行解密处理以获取所述目标图片和所述图片信息。通过在传输过程中采用预先约定的密钥对图片和图片信息进行加解密,可以保证图片和图片信息在传输过程中的安全性。
结果获取单元12,用于基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
具体实现中,所述结果获取单元12可以基于所述图片信息获取所述目标图片中的图片元素,所述图片元素可以包括所述目标图片的地域、时间、气候、格式、物体等图片元素,所述结果获取单元12可以获取所述图片元素对应的分类结果,进一步的,所述结果获取单元12可以在所训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素,可以理解的是,所述图片分类装置1可以预先对多个历史图片进行训练,并在训练后生成历史图片元素集合,所述历史图片为所述图片分类应用的开发人员所收集的多个图片,通过对历史图片进行训练,可以提升图片识别的准确性。所述结果获取单元12分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,所述结果获取单元12可以基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
具体的,请一并参见图4,为本发明实施例提供了结果获取单元的结构示意图。如图4所示,所述结果获取单元12可以包括:
第一元素获取子单元121,用于采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;
具体实现中,所述图片元素可以包括所述目标图片的地域、时间、气候、格式、物体等图片元素,所述第一元素获取子单元121可以采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素,所述预设图片识别方式可以包括以下至少一个识别方式:基于所述地域信息识别所述目标图片的地域图片元素;基于所述时间信息识别所述目标图片的时间图片元素;基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;基于所述图片格式信息识别所述目标图片的格式图片元素;基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
所述第一元素获取子单元121,还用于在预先训练的历史图片元素集合中获取与所述图 片元素相关联的至少一个历史图片元素;
相似度计算子单元122,用于分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
结果生成子单元123,用于基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果;
具体实现中,所述第一元素获取子单元121可以在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素,可以理解的是,所述图片分类装置1可以预先对多个历史图片进行训练,并在训练后生成历史图片元素集合,所述历史图片为所述图片分类对应的开发人员所收集的多个图片,通过对历史图片进行训练,可以提升图片识别的准确性。所述相似度计算子单元122分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,例如:针对所述目标图片的物体图片元素,可以理解的是,物体图片元素是基于简略化处理的所述目标图片所识别的,因此其为物体的边缘化形状,所述第一元素获取子单元121可以获取与该物体图片元素相似的至少一个历史图片元素,所述相似度计算子单元122分别计算与该物体图片元素的形状的元素相似度。所述结果生成子单元123可以基于所述各历史图片元素以及所述各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
文本生成单元13,用于对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本;
具体实现中,所述文本生成单元13可以对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本,所述分类文本用于描述针对需要进行分类的所述目标图片的具体分类类别,例如:风景类、人物类、气候类等,并且所述分类文本可以被所述用户终端所识别,以对所述目标图片进行进一步的分类处理。
具体的,请一并参见图5,为本发明实施例提供了文本生成单元的结构示意图。如图5所示,所述文本生成单元13可以包括:
第二元素获取子单元131,用于在所述至少一个历史图片元素中获取第一历史图片元素,第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
文本生成子单元132,用于确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本;
具体实现中,所述图片分类装置1可以预先设置有相似度阈值,所述第二元素获取子单元131在所述至少一个历史图片元素中获取元素相似度大于预设相似度阈值的第一历史图 片元素,所述文本生成子单元132可以确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。其中,所述分类文本用于描述针对需要进行分类的所述目标图片的具体分类类别,例如:风景类、人物类、气候类等,并且所述分类文本可以被所述用户终端所识别,以对所述目标图片进行进一步的分类处理。
文本发送单元14,用于将所述分类文本发送至所述用户终端;
具体实现中,所述文本发送单元14可以将所述分类文本发送至所述用户终端,优选的,所述文本发送单元14可以采用上述约定的所述密钥对所述分类文本进行加密处理,并采用多路传输方式发送至所述用户终端,所述用户终端在获取到所述分类文本时,可以对其进行解密处理,并按照所述分类文本对所述目标图片进行分类输出,进一步的,所述用户终端可以将所述目标图片存放至所述分类文本所指示的分类文件夹中,可以理解的是,一个图片可以存放于多个分类文件夹中,当检测到用户对所述目标图片所在的分类文件夹的打开操作时,所述用户终端可以在该分类文件夹下对所述目标图片进行输出显示。
在本发明实施例中,通过获取用户终端传输的需要进行分类的目标图片以及该目标图片关联的图片信息,可以基于图片信息获取目标图片中的图片元素,获取图片元素对应的分类结果,并对分类结果进行筛选处理以生成该目标图片对应的分类文本,最终将分类文本发送至用户终端,以使用户终端按照该分类文本对该目标图片进行分类输出。通过获取图片关联的图片信息,并依据图片信息获取的图片元素对图片进行相应分类,实现了对图片进行分类的过程,保证了对图片的管理效果,由于对图片进行了分类处理,可以依据分类快速定位用户终端中的图片,提升了图片浏览的效率;通过采用多路传输方式对图片、图片信息及分类文本进行传输,可以避免传输过程中受到网络环境的影响,保证了图片、图片信息及分类文本传输的稳定性,同时可以提升图片、图片信息及分类文本的传输效率;通过在传输过程中采用预先约定的密钥对图片、图片信息及分类文本进行加解密,可以保证图片、图片信息及分类文本在传输过程中的安全性;通过加入地域信息对图片进行分类,丰富了对图片进行分类的类别,提升了对图片进行分类的准确性;通过对历史图片进行训练,可以提升图片识别的准确性。
请参见图6,为本发明实施例提供了一种电子设备的结构示意图。如图6所示,本发明实施例的所述电子设备可以包括:壳体600、电路板601、处理器602、通信总线603、输入设备604、输出设备605和存储器606,其中,电路板601安置在壳体600围成的空间内部,处理器602、通信总线603、存储器606、输入设备604和输出设备605设置在电路板601上;通信总线603用于实现这些组件之间的连接通信。其中,存储器606可以是高速RAM 存储器,也可以是非不稳定的存储器(non-volatile memory),例如至少一个磁盘存储器。存储器606可选的还可以是至少一个位于远离前述处理器602的存储装置。其中,处理器602可以结合图3-图5所描述的图片分类装置,存储器606中存储可执行程序代码;且处理器602通过读取存储器606中存储的可执行程序代码来运行与可执行程序代码对应的程序,以用于执行以下步骤:
控制输入设备604获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本;
控制输出设备605将所述分类文本发送至所述用户终端,以使得所述用户终端识别所述分类文本以对所述目标图片进行分类输出。
在可选的实施例中,所述处理器602调用存储器606中存储的可执行程序代码执行控制输入设备604获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息时,具体执行以下操作:
控制输入设备604获取用户终端采用多路传输方式所传输的待分类的目标图片以及与所述目标图片相关联的图片信息。
在可选的实施例中,所述处理器602调用存储器606中存储的可执行程序代码执行基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果时,具体执行以下操作:
采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;
在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
基于所述各至少一个历史图片元素中历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
在可选的实施例中,所述图片信息包括地域信息、时间信息、图片格式信息、简略化处理的所述目标图片;
所述预设图片识别方式包括以下至少一个识别方式:
基于所述地域信息识别所述目标图片的地域图片元素;
基于所述时间信息识别所述目标图片的时间图片元素;
基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;
基于所述图片格式信息识别所述目标图片的格式图片元素;
基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
在可选的实施例中,所述处理器602调用存储器606中存储的可执行程序代码执行对所述分类结果进行筛选处理,以生成所述目标图片对应的分类文本时,具体执行以下操作:
在所述至少一个历史图片元素中获取第一历史图片元素,第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。
在本发明实施例中,通过获取用户终端传输的需要进行分类的目标图片以及该目标图片关联的图片信息,可以基于图片信息获取目标图片中的图片元素,获取图片元素对应的分类结果,并对分类结果进行筛选处理以生成该目标图片对应的分类文本,最终将分类文本发送至用户终端,以使用户终端按照该分类文本对该目标图片进行分类输出。通过获取图片关联的图片信息,并依据图片信息获取的图片元素对图片进行相应分类,实现了对图片进行分类的过程,保证了对图片的管理效果,由于对图片进行了分类处理,可以依据分类快速定位用户终端中的图片,提升了图片浏览的效率;通过采用多路传输方式对图片、图片信息及分类文本进行传输,可以避免传输过程中受到网络环境的影响,保证了图片、图片信息及分类文本传输的稳定性,同时可以提升图片、图片信息及分类文本的传输效率;通过加入地域信息对图片进行分类,丰富了对图片进行分类的类别,提升了对图片进行分类的准确性;通过对历史图片进行训练,可以提升图片识别的准确性。
本发明实施例还提供了一种存储介质,其中,所述存储介质用于存储应用程序,所述应用程序用于在运行时执行上述提供的图片分类方法。
本发明实施例还提供了一种应用程序,其中,所述应用程序用于在运行时执行上述提供的图片分类方法。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,所述的程序可存储于一计算机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。其中,所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory,ROM)或随机存储记忆体(Random Access Memory,RAM)等。
以上所揭露的仅为本发明较佳实施例而已,当然不能以此来限定本发明之权利范围,因此依本发明权利要求所作的等同变化,仍属本发明所涵盖的范围。

Claims (17)

  1. 一种图片分类方法,其特征在于,包括:
    获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
    基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
    对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本,所述分类文本用于被所述用户终端识别,并对所述目标图片进行分类输出;
    将所述分类文本发送至所述用户终端。
  2. 根据权利要求1所述的方法,其特征在于,所述获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息,包括:
    通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息。
  3. 根据权利要求1或2所述的方法,其特征在于,所述基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果,包括:
    采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;
    在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
    分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
    基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
  4. 根据权利要求3所述的方法,其特征在于,所述图片信息包括地域信息、时间信息、图片格式信息、简略化处理的所述目标图片;
    所述预设图片识别方式包括以下至少一个识别方式:
    基于所述地域信息识别所述目标图片的地域图片元素;
    基于所述时间信息识别所述目标图片的时间图片元素;
    基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;
    基于所述图片格式信息识别所述目标图片的格式图片元素;
    基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
  5. 根据权利要求3或4所述的方法,其特征在于,所述对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本,包括:
    在所述至少一个历史图片元素中获取第一历史图片元素,其中所述第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
    确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。
  6. 一种图片分类装置,其特征在于,包括:
    信息获取单元,用于获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
    结果获取单元,用于基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
    文本生成单元,用于对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本;
    文本发送单元,用于将所述分类文本发送至所述用户终端,所述分类文本用于被所述用户终端识别,并对所述目标图片进行分类输出。
  7. 根据权利要求6所述的装置,其特征在于,所述信息获取单元具体用于通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息。
  8. 根据权利要求6或7所述的装置,其特征在于,所述结果获取单元包括:
    第一元素获取子单元,用于采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;且在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
    相似度计算子单元,用于分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
    结果生成子单元,用于基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对 应的分类结果。
  9. 根据权利要求8所述的装置,其特征在于,所述图片信息包括地域信息、时间信息、图片格式信息、简略化处理的所述目标图片;
    所述预设图片识别方式包括以下至少一个识别方式:
    基于所述地域信息识别所述目标图片的地域图片元素;
    基于所述时间信息识别所述目标图片的时间图片元素;
    基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;
    基于所述图片格式信息识别所述目标图片的格式图片元素;
    基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
  10. 根据权利要求8或9所述的方法,其特征在于,所述文本生成单元包括:
    第二元素获取子单元,用于在所述至少一个历史图片元素中获取第一历史图片元素,其中所述第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
    文本生成子单元,用于确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。
  11. 一种电子设备,其特征在于,包括:壳体、处理器、存储器、电路板、输入设备和输出设备,其中,电路板安置在壳体围成的空间内部,处理器、存储器、输入设备和输出设备设置在电路板上;存储器用于存储可执行程序代码;处理器通过读取存储器中存储的可执行程序代码来运行与可执行程序代码对应的程序,以用于执行以下步骤:
    控制输入设备获取用户终端传输的待分类的目标图片以及与所述目标图片相关联的图片信息;
    基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果;
    对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本;
    控制输出设备将所述分类文本发送至所述用户终端,以使得所述用户终端识别所述分类文本以对所述目标图片进行分类输出。
  12. 根据权利要求11所述的电子设备,其特征在于,所述处理器用于:
    通过多路传输方式获取所述用户终端传输的待分类的目标图片以及与所述目标图片相 关联的图片信息。
  13. 根据权利要求11或12所述的电子设备,其特征在于,所述处理器用于通过以下步骤基于所述图片信息获取所述目标图片中的图片元素,并获取所述图片元素对应的分类结果:
    采用预设图片识别方式,并基于所述图片信息获取所述目标图片中的图片元素;
    在预先训练的历史图片元素集合中获取与所述图片元素相关联的至少一个历史图片元素;
    分别计算所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度;
    基于所述至少一个历史图片元素中各历史图片元素以及所述至少一个历史图片元素中各历史图片元素与所述图片元素间的元素相似度,生成所述图片元素对应的分类结果。
  14. 根据权利要求13所述的电子设备,其特征在于,所述图片信息包括地域信息、时间信息、图片格式信息、简略化处理的所述目标图片;
    所述预设图片识别方式包括以下至少一个识别方式:
    基于所述地域信息识别所述目标图片的地域图片元素;
    基于所述时间信息识别所述目标图片的时间图片元素;
    基于所述地域信息和所述时间信息识别所述目标图片的气候图片元素;
    基于所述图片格式信息识别所述目标图片的格式图片元素;
    基于所述简略化处理的所述目标图片,识别所述目标图片的物体图片元素。
  15. 根据权利要求13或14所述的电子设备,其特征在于,所述处理器用于通过以下步骤对所述分类结果进行筛选处理,并在筛选处理后生成所述目标图片对应的分类文本:
    在所述至少一个历史图片元素中获取第一历史图片元素,其中所述第一历史图片元素与所述图片元素间的元素相似度大于预设相似度阈值;
    确定所述第一历史图片元素对应的分类类别,并基于所述分类类别生成所述目标图片对应的分类文本。
  16. 一种存储介质,用于存储应用程序,所述应用程序用于在运行时执行如权利要求1-5中任一项所述的图片分类方法。
  17. 一种应用程序,当其被运行时,执行如权利要求1-5中任一项所述的图片分类方法。
PCT/CN2016/088228 2015-12-23 2016-07-01 一种图片分类方法、装置及电子设备 WO2017107431A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510977767.4 2015-12-23
CN201510977767.4A CN105608189B (zh) 2015-12-23 2015-12-23 一种图片分类方法、装置及电子设备

Publications (1)

Publication Number Publication Date
WO2017107431A1 true WO2017107431A1 (zh) 2017-06-29

Family

ID=55988128

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/088228 WO2017107431A1 (zh) 2015-12-23 2016-07-01 一种图片分类方法、装置及电子设备

Country Status (2)

Country Link
CN (1) CN105608189B (zh)
WO (1) WO2017107431A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117576247A (zh) * 2024-01-17 2024-02-20 江西拓世智能科技股份有限公司 一种基于人工智能的图片生成方法及系统

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105608189B (zh) * 2015-12-23 2019-10-11 北京金山安全软件有限公司 一种图片分类方法、装置及电子设备
CN106372471B (zh) * 2016-08-31 2019-05-14 宇龙计算机通信科技(深圳)有限公司 一种ar注册方法及系统
CN109213862B (zh) * 2018-08-21 2020-11-24 北京京东尚科信息技术有限公司 物体识别方法和装置、计算机可读存储介质
CN110032887A (zh) * 2019-02-27 2019-07-19 努比亚技术有限公司 一种图片隐私保护方法、终端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
CN104133899A (zh) * 2014-08-01 2014-11-05 百度在线网络技术(北京)有限公司 图片搜索库的生成方法和装置、图片搜索方法和装置
US20150269143A1 (en) * 2014-03-21 2015-09-24 Samsung Techwin Co., Ltd. Image processing system and method
CN105159959A (zh) * 2015-08-20 2015-12-16 广东欧珀移动通信有限公司 图像文件的处理方法和系统
CN105608189A (zh) * 2015-12-23 2016-05-25 北京金山安全软件有限公司 一种图片分类方法、装置及电子设备
CN105653152A (zh) * 2015-12-23 2016-06-08 北京金山安全软件有限公司 一种图片处理方法、装置及电子设备

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108073948A (zh) * 2012-01-17 2018-05-25 华为技术有限公司 一种照片分类管理方法、服务器、装置及系统
CN104077312A (zh) * 2013-03-28 2014-10-01 腾讯科技(深圳)有限公司 一种图片分类方法和装置
CN103530114B (zh) * 2013-09-27 2017-01-18 贝壳网际(北京)安全技术有限公司 一种图片管理方法和装置
CN104850547B (zh) * 2014-02-13 2019-06-25 腾讯科技(深圳)有限公司 图片展示方法及装置
CN104331499A (zh) * 2014-11-19 2015-02-04 希姆通信息技术(上海)有限公司 移动终端及其图片分类装置及方法
CN104536990B (zh) * 2014-12-10 2018-03-27 广东欧珀移动通信有限公司 一种图片显示方法及终端

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150269143A1 (en) * 2014-03-21 2015-09-24 Samsung Techwin Co., Ltd. Image processing system and method
CN104133899A (zh) * 2014-08-01 2014-11-05 百度在线网络技术(北京)有限公司 图片搜索库的生成方法和装置、图片搜索方法和装置
CN105159959A (zh) * 2015-08-20 2015-12-16 广东欧珀移动通信有限公司 图像文件的处理方法和系统
CN105608189A (zh) * 2015-12-23 2016-05-25 北京金山安全软件有限公司 一种图片分类方法、装置及电子设备
CN105653152A (zh) * 2015-12-23 2016-06-08 北京金山安全软件有限公司 一种图片处理方法、装置及电子设备

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117576247A (zh) * 2024-01-17 2024-02-20 江西拓世智能科技股份有限公司 一种基于人工智能的图片生成方法及系统
CN117576247B (zh) * 2024-01-17 2024-03-29 江西拓世智能科技股份有限公司 一种基于人工智能的图片生成方法及系统

Also Published As

Publication number Publication date
CN105608189A (zh) 2016-05-25
CN105608189B (zh) 2019-10-11

Similar Documents

Publication Publication Date Title
WO2017107431A1 (zh) 一种图片分类方法、装置及电子设备
US9436690B2 (en) System and method for predicting a geographic origin of content and accuracy of geotags related to content obtained from social media and other content providers
TWI583191B (zh) 相片及視頻搜尋
CN105518679B (zh) 图片管理方法以及图片同步方法
CN102591868B (zh) 用于拍照指南自动生成的系统和方法
US20200112557A1 (en) Systems, methods and apparatuses for identity access management and web services access
US20160267144A1 (en) Collecting and generating geo-tagged social media data through a network router interface
KR20160074500A (ko) 모바일 비디오 서치 기법
US20140250457A1 (en) Video analysis system
WO2017113840A1 (zh) 一种信息推荐方法及装置
KR20110096500A (ko) 위치 기반의 통신 방법 및 시스템
US11184307B2 (en) System, apparatus, method, and non-transitory computer readable medium for providing location information by transmitting an image including the location information through a chatroom
WO2019052354A1 (zh) 图像处理方法和系统、计算机设备
CN105653152A (zh) 一种图片处理方法、装置及电子设备
CN104869144A (zh) 一种信息分享方法及电子设备
CN104572830A (zh) 推荐拍摄信息的处理方法及装置
CN113869063A (zh) 数据推荐方法、装置、电子设备及存储介质
WO2014043738A1 (en) Facial recognition with social networking aiding
US20130230248A1 (en) Ensuring validity of the bookmark reference in a collaborative bookmarking system
WO2016172968A1 (zh) 传输云端文件的方法、终端及云端服务器
CN106778449B (zh) 动态影像的物件辨识方法及自动截取目标图像的互动式影片建立方法
JP2019220152A (ja) イメージフィルタリング方法、電子機器および記録媒体
KR102407665B1 (ko) 이미지에서 추출된 키워드를 이용하여 이미지를 필터링하기 위한 방법과 시스템 및 비-일시적인 컴퓨터 판독 가능한 기록 매체
US20150302059A1 (en) Content recommendation apparatus and the method thereof
US11374893B2 (en) Methods, servers, and non-transitory computer readable record media for converting image to location data

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: 16877244

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

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 05.10.2018)

122 Ep: pct application non-entry in european phase

Ref document number: 16877244

Country of ref document: EP

Kind code of ref document: A1