CN112199546A - Photo storage management system and method - Google Patents
Photo storage management system and method Download PDFInfo
- Publication number
- CN112199546A CN112199546A CN202011398519.1A CN202011398519A CN112199546A CN 112199546 A CN112199546 A CN 112199546A CN 202011398519 A CN202011398519 A CN 202011398519A CN 112199546 A CN112199546 A CN 112199546A
- Authority
- CN
- China
- Prior art keywords
- unit
- target
- photo
- marking
- storage
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/55—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
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)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a photo storage management system, comprising: the image marking module is used for marking the received photos and marking all contents displayed on the photos, the classifying module is used for detecting the photo contents, dividing the photo contents into three categories according to the difference of the photo contents and storing the three categories in the storage module, and the retrieval module is used for retrieving the photos and comprises searching the photos by using pictures and searching the pictures by using keywords. The invention also relates to a photo storage management method, which receives the photos, marks the content of the photos, stores the photos in a classified manner and realizes quick retrieval. The method and the device have the functions of searching the images by the images and searching the images by the keywords, are associated with the mark content before storage, and enable a user to mark the photo content in a personalized mode to generate the storage space with the individual identity characteristics and the language characteristics.
Description
Technical Field
The invention relates to the technical field of information storage management, in particular to a photo storage management system and a photo storage management method.
Background
More and more terminal users use the mobile terminal as a tool for recording life and capturing beautiful scenery. The continuous push by the industry of large-pixel cell phones has also demonstrated that consumer demand is continuously growing for taking pictures. According to data research conducted by the mobile internet design center in 2012 for cell phone photography, it was shown that 44% of people have become accustomed to processing photographs on cell phones. This ratio is undoubtedly still rising year by year with the advancement of smartphone technology.
However, most users only take pictures of mobile terminals and only take pictures, and rarely deal with sorting. Even if the photos are uploaded to the network, the stored photos are disordered because of no arrangement, when the photos are required to be searched, the photos are often difficult to search, and some photos are too long in time interval, so that the parties cannot remember specific time or location information, the searching is more difficult, the photos in the storage space are often browsed once, the required photos can be found, the searching efficiency is low, and great inconvenience is brought to the use of a user.
Disclosure of Invention
The invention aims to provide a labeling function when a photo is input, store the input photo according to different contents in a classified manner, and provide functions of searching the photo by using a picture and searching the photo by using keywords so as to facilitate the sorting and searching of the photo.
In one aspect of the present application, there is provided a photo storage management system including: the image marking module is used for marking the received photos and marking each content presented on the photos, and comprises an image identification unit, an automatic marking unit, a manual marking unit and a first instruction receiving unit, wherein the automatic marking unit and the manual marking unit are connected with the image identification unit and the first instruction receiving unit; the classification module is used for classifying the received photos and comprises a first target detection network and a second target detection network; the storage module is used for storing the classified photos and comprises a first storage unit, a second storage unit and a third storage unit; the first storage unit corresponds to a first target detection network, and the second storage unit corresponds to a second target detection network; the first storage unit is used for storing a photo comprising a first object/a third object; the second storage unit is used for storing the photo comprising the second object/the fourth object; the third storage unit is used for storing other photos; the retrieval module is used for retrieving the target photo from the storage module, and comprises an object receiving unit, a feature extraction unit, a second instruction receiving unit, a target matching unit and a result output unit, wherein the object receiving unit is sequentially connected with the feature extraction unit and the target matching unit, the second instruction receiving unit is connected with the target matching unit, and the target matching unit is sequentially connected with the result output unit.
Further, the manual marking unit comprises an automatic target identification unit and a manual target framing unit; the automatic target identification unit is connected with the image identification unit and the first instruction receiving unit, and the manual target frame selection unit is connected with the first instruction receiving unit.
Furthermore, the first object is a person, the second object is an object, the third object is a name or a title of the person, and the fourth object is a name or a code number of the object.
In another aspect of the present application, a photo storage management method is provided, which is applied to a photo storage management system, and includes: and receiving the photo, judging whether the photo comprises a first object/a third object through the first target detection network, if so, storing the photo in a first storage unit, otherwise, entering a second target detection network, judging whether the photo comprises a second object/a fourth object through the second target detection network, if so, storing the photo in a second storage unit, otherwise, storing the photo in a third storage unit.
Further, calculating areas occupied by the identifiable objects on the photo, wherein the target objects occupying different areas correspond to different first evaluation values, detecting position coordinates of the identifiable objects on the photo, wherein the target objects positioned at different coordinates correspond to different second evaluation values, and each identifiable object comprehensive area size and position evaluation value corresponds to one comprehensive evaluation value; the first target detection network is provided with a first evaluation threshold value, when the comprehensive evaluation value of the recognizable objects on the photo detected by the first target detection network is larger than the first evaluation threshold value, the photo is stored in the first storage unit, otherwise, the photo enters the second target detection network, the first target detection network is provided with a second evaluation threshold value, when the comprehensive evaluation value of the recognizable objects on the photo detected by the second target detection network is larger than the second evaluation threshold value, the photo is stored in the second storage unit, otherwise, the photo is stored in the third storage unit.
The method further comprises an automatic marking mode, wherein an input photo is received, when the automatic marking unit is selected, the image recognition unit is called to recognize a possible interested target area in the photo, keyword pre-marking is carried out, and pre-marked keywords are selected to be accepted or cancelled according to the content received by the first instruction receiving unit.
The method comprises the steps that a manual marking mode is further included, an input photo is received, when a manual marking unit is selected, an image editing page is entered, when automatic target recognition is selected, an image recognition unit is called, a possible interested target area is selected on the photo, and the possible interested target area is screened and labeled according to the content received by a first instruction receiving unit; when the manual target frame selection is selected, the first instruction receiving unit receives the input marking curve and the marking keywords, and the area surrounded by the marking curve corresponds to the marking keywords.
Further, the method comprises a picture searching mode, wherein the picture searching mode comprises the following steps: the object receiving unit receives a target object image, the feature extraction unit extracts target object features, the target object to be retrieved is determined according to the first information received by the second instruction receiving unit, the type of the target object is judged according to the target object features to be retrieved, the corresponding storage unit is selected according to the type of the target object, a pre-comparison photo group meeting retrieval conditions is output according to the second information of the second instruction receiving unit, the target matching unit performs target matching on the target object to be retrieved and the pre-comparison photo group, and finally a photo matched with the target object to be retrieved is output.
Further, the method comprises a searching mode by keywords, wherein the searching mode by keywords comprises the following steps: the object receiving unit receives the keywords, the feature extraction unit extracts keyword information, determines a target object to be retrieved according to the first information received by the second instruction receiving unit, judges the category of the target object according to the features of the target object to be retrieved, and selects a corresponding storage unit according to the category of the target object; and outputting a pre-comparison picture group meeting the retrieval condition according to the second information of the second instruction receiving unit, matching the keyword to be retrieved with the mark keyword on the pre-comparison picture group by the target matching unit, and outputting the picture matched with the keyword to be retrieved.
Compared with the prior art, the invention has the following substantial effects: (1) the automatic marking/manual marking function is provided, so that a user can mark the photo information at the initial storage stage conveniently, and the photo information can be stored firstly and then marked; (2) the system is divided into three storage units according to people, objects and others, and the storage units respectively store corresponding target photos, so that the system has higher pertinence in photo retrieval; (3) comprehensively scoring the identifiable objects according to the areas occupied by the identifiable objects and the coordinate positions of the identifiable objects in the photos, storing the identifiable objects according to the types of the identifiable objects with the highest comprehensive scores, and fully combining the purposeful shooting characteristics of the photos; (4) the functions of searching pictures by pictures and searching pictures by keywords are provided, and the functions are associated with the mark content before storage, so that the user can mark the photo content in a personalized way, and a storage space with individual identity characteristics and language characteristics is generated.
Drawings
FIG. 1 is a flow chart of the sorted storage of the present invention;
FIG. 2 is a block diagram of a classification module of the present invention;
FIG. 3 is a block diagram of a memory module of the present invention;
FIG. 4 is a block diagram of the search module of the present invention;
FIG. 5 is a block diagram of an image marking module of the present invention;
FIG. 6 is a flowchart of a graph search;
FIG. 7 is a flowchart of keyword searching.
Detailed Description
The invention will be further described with reference to the accompanying drawings in which:
a photo management system is installed on user terminal equipment and comprises an image marking module, a classification module, a storage module and a retrieval module, wherein the user terminal equipment comprises a mobile phone, an ipad, a computer, a notebook computer and the like. A photo management method receives a photo, as shown in FIG. 1, a first target detection network 210 judges whether the photo includes a first object/a third object, if yes, the photo is stored in a first storage unit, otherwise, the photo enters a second target detection network, the second target detection network judges whether the photo includes a second object/a fourth object, if yes, the photo is stored in a second storage unit, otherwise, the photo is stored in a third storage unit. The following describes a photo management system and method. The image marking module is configured to mark the received photo, and mark each content presented on the photo, as shown in fig. 2, and includes an image recognition unit 110, an automatic marking unit 120, a manual marking unit 140, and a first instruction receiving unit 130, where the automatic marking unit 120 and the manual marking unit 140 are both connected to the image recognition unit 110 and the first instruction receiving unit 130. The pictures in the application refer to marked pictures or stored pictures. The image marking module comprises two marking modes, namely an automatic marking mode and a manual marking mode, a user can automatically select to mark through automatic marking or manual marking, or automatic marking can be carried out firstly, and manual marking is carried out again if the result of automatic marking is not satisfactory. The automatic marking mode is suitable for the same target object which is stored on the user terminal equipment and is marked with words once, and the manual marking mode is suitable for the target object which is received for the first time.
The user terminal equipment receives an input photo and then selects to enter the image marking module, when the user selects to enter the image marking module and selects the automatic marking mode, the image recognition unit 110 is called to intelligently recognize the content of the photo, the intelligently recognized marking target which is possibly interested by the user is displayed on the photo, pre-marking is carried out through the automatic marking unit 120, namely, a marking keyword corresponding to the marking target is generated, and the user can select to accept the marking target or cancel the marking target, or accept the marking target and cancel the marking keyword, and re-input the marking keyword. The terminal device receives instruction information from the user through the first instruction receiving unit 130, for example, selecting "confirm" or "cancel", typing in a labeling keyword.
When a user selects to enter the image marking module, when the user selects a manual marking mode, automatic target identification or manual target framing can be selected, wherein the automatic target identification refers to calling the image identification unit 110 to identify a labeled target which may be interested by the user, and the user manually screens and labels the labeled target; the manual target frame selection refers to that a user identifies an interested area on a terminal equipment interface through a marking curve and marks keywords, and the area surrounded by the marking curve corresponds to the marked keywords.
When the automatic target recognition 141 is selected, the image recognition unit 110 is called to select a possible target area of interest on the picture, the possible target area of interest is screened and labeled according to the content received by the first instruction receiving unit 130, and when the manual target frame selection 142 is selected, the first instruction receiving unit 130 receives the input marking curve and the labeling keyword, and the area surrounded by the marking curve corresponds to the labeling keyword. When the annotation is carried out, the user can set the keyword annotation with the language characteristic thereof which is adaptive to the individual identity according to the personal preference.
As shown in FIG. 3, the classification module classifies the received photograph including a first object detection network 210 and a second object detection network 220. As shown in fig. 4, the memory module includes a first memory cell 310, a second memory cell 320, and a third memory cell 330. The first storage unit corresponds to a first object detection network, the second storage unit corresponds to a second object detection network, and the first storage unit is used for storing a photo comprising a first object/a third object; the second storage unit is used for storing the photo comprising the second object/the fourth object; the third storage unit is used for storing other photos.
The photos marked by the image marking module enter the classification module and the storage module, and a user can also choose to skip the image marking module and directly enter the classification module and the storage module. The first target detection network and the second target detection network are obtained by deep learning training and can detect image information or character information on a picture, whether the picture comprises a first object/a third object or not is judged through the first target detection network, if yes, the picture is stored in the first storage unit, otherwise, the picture enters the second target detection network, whether the picture comprises a second object/a fourth object or not is judged through the second target detection network, if yes, the picture is stored in the second storage unit, and if not, the picture is stored in the third storage unit.
And the pictures passing through the image marking module are subjected to character recognition after entering the classification module, and are classified according to standard keywords. If the recognized characters belong to the third object, the recognized characters are stored in the first storage unit, if the recognized characters belong to the fourth object, the recognized characters are stored in the second storage unit, and if not, the recognized characters are stored in the third storage unit. The third object is a name or a title, the fourth object is a name or a code, and the names, the titles, the names or the codes are pre-stored in the first target detection network 210 and the second target detection network.
The photos which do not directly enter the classification module through the image marking module are subjected to image information identification, and classification is carried out according to the identified image information. In the process of target detection, different from a common target detection output large number of pre-selection frames, because the picture is based on the shooting of the target, the shot target often occupies a key position (upper left, lower left, upper right, lower right or central area) of the whole picture, and the size is at least 1/9 of the whole picture, and the key position and size threshold value can be set autonomously. In order to improve the accuracy and rapidity of detection, detection algorithms are set in the first target detection network 210 and the second target detection network, and the position coordinates and the occupied area of the target object on the photo are integrated to judge the main object type of the photo.
Calculating the areas occupied by the identifiable objects on the picture, wherein the target objects occupying different areas correspond to different first evaluation values, detecting the position coordinates of the identifiable objects on the picture, comparing the target objects positioned at different coordinates with different second evaluation values, and each identifiable object has a comprehensive area size and a comprehensive position corresponding to a comprehensive evaluation value. Recognizable objects having a comprehensive evaluation value smaller than a certain threshold value do not appear, and the recognizable objects appearing on the screen are determined according to the magnitude of the comprehensive evaluation value, and the number of the appearance can be set. Further, the first object detection network 210 is provided with a first evaluation threshold value, and when the integrated evaluation value of the position coordinates and the occupied area size of the identifiable object detected by the first object detection network 210 is greater than the first evaluation threshold value, the identifiable object is stored in the first storage unit, otherwise, the second object detection network is entered, and the first object detection network 210 is provided with a second evaluation threshold value, and when the integrated evaluation value of the position coordinates and the occupied area size of the identifiable object detected by the second object detection network is greater than the second evaluation threshold value, the identifiable object is stored in the second storage unit, otherwise, the identifiable object is stored in the third storage unit.
It can be simply understood that if an object occupying the largest area of the entire photograph and located closer to the center of the photograph than other objects is a first object, it is stored in the first storage unit, if an object occupying the largest area of the entire photograph and located closer to the center of the photograph than other objects is a second object, it is stored in the second storage unit, otherwise, it is stored in the third storage unit. Wherein the first object is a person and the second object is an object. The transfer of the marking and storage locations can also be done manually, in case the user is not satisfied with the results of the automatic classification.
As shown in fig. 5, the retrieval module includes an object receiving unit 411, a feature extracting unit 412, a second instruction receiving unit 420, a target matching unit 430, and a result outputting unit 440, where the object receiving unit 411 is sequentially connected to the feature extracting unit 412 and the target matching unit 430, the second instruction receiving unit 420 is connected to the target matching unit 430, and the target matching unit 430 is sequentially connected to the result outputting unit 440. When the image is searched, two modes of searching images by using images and searching images by using keywords are included.
Searching the images according to the images: as shown in fig. 6, a target object graph is input first, the object receiving unit receives a target object, the feature extraction unit extracts features of the target object, determines a target object to be retrieved according to first information received by the second instruction receiving unit, determines a category of the target object according to the features of the target object to be retrieved, and selects a corresponding storage unit according to the category of the target object; and outputting a pre-comparison picture group meeting the retrieval condition according to the second information of the second instruction receiving unit, performing target matching on the target object to be retrieved and the pre-comparison picture group by the target matching unit (comparing the target object to be retrieved with the target area of interest marked on each picture), and outputting a picture matched with the target object to be retrieved. The first information refers to an externally input confirmation instruction for confirming which object to be retrieved is, and the second information refers to an externally input retrieval condition, such as a time span.
For example, a graph is input, a house and a person exist on the graph, all photos including the person on the graph within 1-6 months of 20 years are required to be retrieved, firstly, which target objects (the house and the person) -the image of the Xiaonan ' included on the input graph are extracted, then, what the target objects to be retrieved are is determined according to the first information received by the second instruction receiving unit (the image of the Xiaonan ' is selected), a storage unit (a first storage unit) corresponding to the category of the target objects (the first object) is determined, then, according to the set retrieval condition, namely, the second information (1-6 months of 20 years) of the second instruction receiving unit, a pre-comparison image group (including all images close to the image of the Xiaonan ') within the time period of 20 years 1-6 months in the first storage unit is output, the input images and the pre-comparison image group are compared one by one, and finally outputting the target picture matched with the Xiaonan.
Searching the images by keywords: as shown in fig. 7, firstly, a keyword is input, the object receiving unit receives the keyword, the feature extraction unit extracts information of the keyword, determines a target object to be retrieved according to the first information received by the second instruction receiving unit, determines a category of the target object according to the feature of the target object to be retrieved, and selects a corresponding storage unit according to the category of the target object; and outputting a pre-comparison picture group meeting the retrieval condition according to the second information of the second instruction receiving unit, matching the keyword to be retrieved with the mark keyword on the pre-comparison picture group by the target matching unit, and outputting the picture matched with the keyword to be retrieved. The first information refers to an externally input confirmation instruction for confirming which object to be retrieved is, and the second information refers to an externally input retrieval condition, such as a time span.
For example, a keyword "xianan" is input, the object receiving unit receives the keyword, the feature extraction unit extracts features of the target object, at this time, a preselection frame may jump out of mom of xianan and an automobile of xianan, the target object is determined to be the "xianan" according to the first information "xianan" received by the second instruction receiving unit, the category is the "third object", a corresponding storage unit (first storage unit) is selected, a pre-comparison image group marked with the "xianan" and meeting second information (20 years and 1 month) of the second instruction receiving unit in the first storage unit is output according to second information of the second instruction receiving unit, namely retrieval conditions (20 years and 1 month) input by a user, the keyword is compared with a labeling keyword of the pre-comparison image group, and an image corresponding to the labeling keyword consistent with the keyword is output. If the input keywords are not stored in the system in advance, the corresponding target objects are not retrieved by default.
The foregoing illustrates and describes the principles, general features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed.
Claims (9)
1. A photograph storage management system, comprising:
the image marking module is used for marking the received photos and marking each content presented on the photos, and comprises an image identification unit, an automatic marking unit, a manual marking unit and a first instruction receiving unit, wherein the automatic marking unit and the manual marking unit are connected with the image identification unit and the first instruction receiving unit;
the classification module is used for classifying the received photos and comprises a first target detection network and a second target detection network;
the storage module is used for storing the classified photos and comprises a first storage unit, a second storage unit and a third storage unit; the first storage unit corresponds to a first target detection network, and the second storage unit corresponds to a second target detection network; the first storage unit is used for storing a photo comprising a first object/a third object; the second storage unit is used for storing the photo comprising the second object/the fourth object; the third storage unit is used for storing other photos;
the retrieval module is used for retrieving the target photo from the storage module, and comprises an object receiving unit, a feature extraction unit, a second instruction receiving unit, a target matching unit and a result output unit, wherein the object receiving unit is sequentially connected with the feature extraction unit and the target matching unit, the second instruction receiving unit is connected with the target matching unit, and the target matching unit is sequentially connected with the result output unit.
2. The photo storage management system of claim 1, wherein the manual tagging unit comprises an automatic object recognition unit and a manual object framing unit; the automatic target identification unit is connected with the image identification unit and the first instruction receiving unit, and the manual target frame selection unit is connected with the first instruction receiving unit.
3. The system of claim 1, wherein the first object is a person, the second object is an object, the third object is a person name or title, and the fourth object is a object name or code number.
4. A photograph storage management method applied to the photograph storage management system according to any one of claims 1 to 3, characterized by comprising: and receiving the photo, judging whether the photo comprises a first object/a third object through the first target detection network, if so, storing the photo in a first storage unit, otherwise, entering a second target detection network, judging whether the photo comprises a second object/a fourth object through the second target detection network, if so, storing the photo in a second storage unit, otherwise, storing the photo in a third storage unit.
5. The photo storage management method according to claim 4, wherein areas occupied by the recognizable objects on the photo are calculated, the target objects occupying different areas correspond to different first evaluation values, the position coordinates where the recognizable objects are located on the photo are detected, the target objects located at different coordinates correspond to different second evaluation values, and one integrated evaluation value corresponds to each of the sizes of the integrated areas and the position evaluation values of the recognizable objects;
the first target detection network is provided with a first evaluation threshold value, when the comprehensive evaluation value of the recognizable objects on the photo detected by the first target detection network is larger than the first evaluation threshold value, the photo is stored in the first storage unit, otherwise, the photo enters the second target detection network, the first target detection network is provided with a second evaluation threshold value, when the comprehensive evaluation value of the recognizable objects on the photo detected by the second target detection network is larger than the second evaluation threshold value, the photo is stored in the second storage unit, otherwise, the photo is stored in the third storage unit.
6. The method of claim 4, comprising an automatic tagging mode, wherein the method receives an input photo, calls the image recognition unit when the automatic tagging unit is selected, recognizes a possible target region of interest in the photo, pre-tags the keyword, and selects to accept or cancel the pre-tagged keyword according to the content received by the first instruction receiving unit.
7. The photo storage management method of claim 4, comprising a manual tagging mode, wherein the manual tagging mode receives an input photo, enters an image editing page when the manual tagging unit is selected, calls the image recognition unit when the automatic target recognition is selected, frames a possible target area of interest on the photo, and filters and labels the possible target area of interest according to the content received by the first instruction receiving unit; when the manual target frame selection is selected, the first instruction receiving unit receives the input marking curve and the marking keywords, and the area surrounded by the marking curve corresponds to the marking keywords.
8. The method of claim 4, comprising a graph search mode, wherein the graph search mode comprises: the object receiving unit receives a target object image, the feature extraction unit extracts target object features, the target object to be retrieved is determined according to the first information received by the second instruction receiving unit, the type of the target object is judged according to the target object features to be retrieved, the corresponding storage unit is selected according to the type of the target object, a pre-comparison photo group meeting retrieval conditions is output according to the second information of the second instruction receiving unit, the target matching unit performs target matching on the target object to be retrieved and the pre-comparison photo group, and finally a photo matched with the target object to be retrieved is output.
9. The method for managing storage of photos according to claim 6 or 7, comprising a keyword search mode, said keyword search mode comprising: the object receiving unit receives the keywords, the feature extraction unit extracts keyword information, determines a target object to be retrieved according to the first information received by the second instruction receiving unit, judges the category of the target object according to the features of the target object to be retrieved, and selects a corresponding storage unit according to the category of the target object; and outputting a pre-comparison picture group meeting the retrieval condition according to the second information of the second instruction receiving unit, matching the keyword to be retrieved with the mark keyword on the pre-comparison picture group by the target matching unit, and outputting the picture matched with the keyword to be retrieved.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011398519.1A CN112199546A (en) | 2020-12-04 | 2020-12-04 | Photo storage management system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011398519.1A CN112199546A (en) | 2020-12-04 | 2020-12-04 | Photo storage management system and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112199546A true CN112199546A (en) | 2021-01-08 |
Family
ID=74033756
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011398519.1A Pending CN112199546A (en) | 2020-12-04 | 2020-12-04 | Photo storage management system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112199546A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104021384A (en) * | 2014-06-30 | 2014-09-03 | 深圳市创冠智能网络技术有限公司 | Face recognition method and device |
CN105279273A (en) * | 2015-10-28 | 2016-01-27 | 广东欧珀移动通信有限公司 | Photo classification method and device |
CN105913052A (en) * | 2016-06-08 | 2016-08-31 | Tcl集团股份有限公司 | Photograph classification management method and system thereof |
CN107526769A (en) * | 2017-07-10 | 2017-12-29 | 北京百度网讯科技有限公司 | Photo processing method and device, equipment and computer-readable recording medium based on artificial intelligence |
CN110046266A (en) * | 2019-03-28 | 2019-07-23 | 广东紫晶信息存储技术股份有限公司 | A kind of intelligent management and device of photo |
CN111295669A (en) * | 2017-06-16 | 2020-06-16 | 马克波尔公司 | Image processing system |
-
2020
- 2020-12-04 CN CN202011398519.1A patent/CN112199546A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104021384A (en) * | 2014-06-30 | 2014-09-03 | 深圳市创冠智能网络技术有限公司 | Face recognition method and device |
CN105279273A (en) * | 2015-10-28 | 2016-01-27 | 广东欧珀移动通信有限公司 | Photo classification method and device |
CN105913052A (en) * | 2016-06-08 | 2016-08-31 | Tcl集团股份有限公司 | Photograph classification management method and system thereof |
CN111295669A (en) * | 2017-06-16 | 2020-06-16 | 马克波尔公司 | Image processing system |
CN107526769A (en) * | 2017-07-10 | 2017-12-29 | 北京百度网讯科技有限公司 | Photo processing method and device, equipment and computer-readable recording medium based on artificial intelligence |
CN110046266A (en) * | 2019-03-28 | 2019-07-23 | 广东紫晶信息存储技术股份有限公司 | A kind of intelligent management and device of photo |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8312374B2 (en) | Information processing apparatus and method and computer program | |
US9042646B2 (en) | Detecting recurring themes in consumer image collections | |
EP2005364B1 (en) | Image classification based on a mixture of elliptical color models | |
JP5170961B2 (en) | Image processing system, image processing apparatus and method, program, and recording medium | |
US8874596B2 (en) | Image processing system and method | |
CN104111989B (en) | The offer method and apparatus of search result | |
JP2007206920A (en) | Image processor and image processing method, retrieving device and method, program and recording medium | |
JP2007513413A (en) | Content recognition for selecting emphasized images | |
CN101930444A (en) | Image search system and method | |
JP2007206919A (en) | Display control device, method, program and storage medium | |
CN102033958A (en) | Photo sort management system and method | |
JP2005510775A (en) | Camera metadata for categorizing content | |
JP2011253424A (en) | Image recognition device and image recognition method and information processing system | |
US20060290789A1 (en) | File naming with optical character recognition | |
US20060026127A1 (en) | Method and apparatus for classification of a data object in a database | |
CN111061887A (en) | News character photo extraction method, device, equipment and storage medium | |
US20230336671A1 (en) | Imaging apparatus | |
KR20210086836A (en) | Image data processing method for searching images by text | |
JP2014092955A (en) | Similar content search processing device, similar content search processing method and program | |
CN107656760A (en) | Data processing method and device, electronic equipment | |
CN112199546A (en) | Photo storage management system and method | |
JP2003330941A (en) | Similar image sorting apparatus | |
CN111159113A (en) | Portable digital photographing equipment system and photo screening method | |
JP2006313497A (en) | Apparatus and method for retrieving image | |
CN113407757B (en) | Image retrieval method and device based on computer |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210108 |