Image compression by object segregation

• US 7,283,672 B1
• Filed: 06/29/2001
• Issued: 10/16/2007
• Est. Priority Date: 06/29/2001
• Status: Expired due to Fees

• Current Assignee: Nagravision SA (Kudelski SA)
• Original Assignee: Scott C Harris
• Inventors: Harris, Scott C.
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US09/681,948
• Time in Patent Office: 2,300 Days
• Field of Search: 382/305, 382/209, 382/217, 382/224, 382/181, 382/203, 707/6, 707/100, 707/200, 715/515, 345/625, 348/91, 358/538
• US Class Current: 382/203
• CPC Class Codes: G06T 9/001 Model-based coding, e.g. wire frame
• CPC Class Codes: H04N 19/27 involving both synthetic and natural picture components, e.g. synthetic natural hybrid coding [SNHC]

Abstract

A system of analyzing and processing images. The image is processed using machine vision techniques to find shapes and objects in the image. In one embodiment, the system finds geometric shapes and textures in the images. In another embodiment, the image processor finds actual objects in the image. Each of these items may be replaced with an indication of the item, along with difference information indicating the difference between the actual object in the image, and the object stored in the database. The difference information may include scale information, rotation information, color information and others. The information in the database may also include sub indicia, such as information about different parts of the item, and/or information about other things that may exist in the image.

Claims

1. A method comprising:

analyzing an electronic file that represents an image to recognize real-life objects within the image; and

replacing recognized real-life objects within the electronic file indicative of the image by an indication representing the real-life object, wherein said replacing includes providing individual part information indicative of how a real-life object within the image differs from a unit real-life object.

2. A method as in claim 1, wherein said individual part information includes information about size and orientation of the actual objects relative to said unit actual objects.

3. An image analyzing device, comprising:

an image obtaining device, obtaining an electronic file indicative of an image;

a database, storing a plurality of image parts representing likely real-life objects which may exist in the image; and

an image processing device, processing said electronic file to recognize actual said real-life objects within said electronic file that correspond to said image parts in said database, and to provide a modified electronic file, indicative of the image, which replaces said recognized real-life objects with indications representing the real life objects based on information in said database, wherein said image processing device also produces additional information that represents how a recognized real-life object within the image differs from real-life objects within the database.

4. A device as in claim 3, wherein said additional information includes information about differences in size and orientation of the recognized part.

5. A device as in claim 3, wherein said image processing device recognizes real-life objects in the image, and finds image parts in said database which correspond to said real-life object.

6. A device as in claim 5, wherein said database also stores information indicative of other objects in said image which may appear near said real-life objects, and wherein said image processing device processes said electronic file to look for said other objects.

7. A method, comprising:

analyzing an image against a database, to find portions of the image which are present in the database, and to replace said portions of the image which are present in the database with information based on said image in the database; and

storing a list of image portions which are not found in the database to be later used to update the database.

8. A method as in claim 7, further comprising sending said list of image portions to a database developer.

9. A method as in claim 7, wherein said analyzing comprises compressing the image using information in the database.

10. A method as in claim 8, further comprising obtaining updates to the database from the database developer.

11. A method comprising:

analyzing an electronic file that represents an image to recognize real-life objects within the image; and

replacing recognized real-life objects within the electronic file indicative of the image by an indication representing the real-life object wherein said real-life objects are each represented by a generic identifier that represents many different objects of the same type, and by individual part information other than said generic identifier representing individual characteristics of the actual object.

12. A method as in claim 11, wherein said individual part information includes information about different species associated with the generic identifier.

13. An image analyzing device, comprising:

an image obtaining device, obtaining an electronic file indicative of an image;

a database, storing a plurality of image parts representing likely real-life objects which may exist in the image; and

an image processing device, processing said electronic file to recognize actual said real-life objects within said electronic file that correspond to said image parts in said database, and to provide a modified electronic file, indicative of the image, which replaces said recognized real-life objects with indications representing the real life objects based on information in said database,

wherein said database stores a generic identifier for each of said actual objects, and individual characteristics for the actual objects which individualize the actual objects.

14. A device as in claim 13, wherein said individual part information includes information about different species associated with the generic identifier.