Method and system for updating images in an image database

• US 8,843,504 B2
• Filed: 03/08/2010
• Issued: 09/23/2014
• Est. Priority Date: 03/09/2009
• Status: Active Grant

• Current Assignee: Alibaba Group Holding Ltd.
• Original Assignee: Alibaba Group Holding Ltd.
• Inventors: Zhou, Chunyi, Yu, Shengyu, Yue, Dajiong
• Primary Examiner(s): Ortiz Ditren, Belix M
• Application Number:
  US12/660,985
• Publication Number:
  US 20100228722A1
• Time in Patent Office: 1,660 Days
• Field of Search: 707/755, 707/802, 707/706, 707/711
• US Class Current: 707/755
• CPC Class Codes: G06F 16/51 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/58 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content

Abstract

Updating images in an image database includes determining an updated image that is among a plurality of images in the image database, extracting a plurality of image features of the updated image and storing a plurality of image feature values that correspond to the plurality of image features, and associating the plurality of image features of the updated image with a search engine, including adding an array comprising the plurality of image feature values to an index of the search engine.

Claims

1. A method for updating an image database, comprising:

detecting, using a processor, an update to an image that is among a plurality of existing images in the image database;

determining an updated image corresponding to the detected update;

extracting a plurality of image features of the updated image and storing a plurality of image feature values that correspond to the plurality of image features, wherein extracting the image features of the updated image includes;

determining whether an image feature table includes a modification time of the updated image;

in the event that the image feature table does not include the modification time of the image, recording a current time as the modification time in the image feature table; and

in the event that the image feature table includes the modification time of the image;

comparing the modification time corresponding to an image identifier of the updated image in an image update table and the image feature table according to a preset schedule; and

recording the plurality of image features in the image feature table; and

associating the plurality of image features of the updated image with a search engine, including adding an array comprising the plurality of image feature values to an index of the search engine.

2. The method of claim 1, wherein determining the updated image includes recording the image identifier of the updated image in the image update table.

3. The method of claim 1, wherein detecting an update includes:

creating a trigger configured to detect an image update in the image database;

setting off the trigger in the event that an UPDATE command is detected in the image database; and

recording the image identifier of the updated image in the image update table.

4. The method of claim 1, further comprising:

recording in the image feature table an image identifier that exists in the image update table but does not exist in the image feature table.

5. The method of claim 1, wherein extracting the image feature of the updated image comprises:

comparing the modification time corresponding to the same image identifier in the image update table and the image feature table according to the preset schedule; and

extracting the feature of the image corresponding to the image identifier, recording the image feature in the image feature table, and updating the current time as the modification time if the modification time in the image feature table is earlier than the modification time in the image update table.

6. The method of claim 1, further comprising:

distributing the plurality of image features of the updated image to a distributed search engine.

14. The method of claim 1, wherein the recording of the plurality of image features includes:

in the event that the modification time corresponding to the image identifier of the updated image in the image update table and the image feature table is the same, performing no action.

7. An image updating system, comprising:

an interface to an image database;

one or more processors coupled to the interface, configured to;

detect an update to an image that is among a plurality of existing images in an image database;

determine an updated image corresponding to the detected update;

extract an image feature of the updated image and storing the image features, wherein extract the image features of the updated image includes;

determine whether an image feature table includes a modification time of the updated image;

in the event that the image feature table does not include the modification time of the image, record a current time as the modification time in the image feature table; and

in the event that the image feature table includes the modification time of the image;

compare the modification time corresponding to an image identifier of the updated image in an image update table and the image feature table according to a preset schedule; and

record the plurality of image features in the image feature table; and

associate the plurality of image features of the updated image with a search engine;

one or more memories coupled to the one or more processors, configured to provide the processors with instructions.

8. The system of claim 7, wherein determining the updated image includes recording the image identifier of the updated image in the image update table.

9. The system of claim 7, wherein detecting an update includes:

creating a trigger configured to detect an image update in the image database;

setting off the trigger in the event that an UPDATE command is detected in the image database; and

recording the image identifier of the updated image in the image update table.

10. The system of claim 7, further comprising:

recording in the image feature table the image identifier that exists in the image update table but does not exist in the image feature table.

11. The system of claim 7, wherein extracting the image feature of the updated image comprises:

comparing the modification time corresponding to the same image identifier in the image update table and an image feature table according to the preset schedule; and

extracting the feature of the image corresponding to the image identifier, recording the image feature in the image feature table, and updating the current time as the modification time if the modification time in the image feature table is earlier than the modification time in the image update table.

12. The system of claim 7, further comprising:

distributing the plurality of image features of the updated image to a distributed search engine.

13. A computer program product for updating an image database, the computer program product being embodied in a computer readable storage medium and comprising computer instructions for:

detecting, using a processor, an update to an image that is among a plurality of existing images in the image database;

determining an updated image corresponding to the detected update;

determining an updated image that is among a plurality of images in the image database;

extracting a plurality of image features of the updated image and storing a plurality of image feature values that correspond to the plurality of image features, wherein extracting the image features of the updated image includes;

determining whether an image feature table includes a modification time of the updated image;

in the event that the image feature table does not include the modification time of the image, recording a current time as the modification time in the image feature table; and

in the event that the image feature table includes the modification time of the image;

comparing the modification time corresponding to an image identifier of the updated image in an image update table and the image feature table according to a preset schedule; and

recording the plurality of image features in the image feature table; and

associating the plurality of image features of the updated image with a search engine, including adding an array comprising the plurality of image feature values to an index of the search engine.