USING MOTION TRACKING AND IMAGE CATEGORIZATION FOR DOCUMENT INDEXING AND VALIDATION

• US 20160342834A1
• Filed: 05/20/2015
• Published: 11/24/2016
• Est. Priority Date: 05/20/2015
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: RAGNET, FRANCOIS, CRAMET, DAMIEN
• Granted Patent: US 9,619,701 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 3/04845 for image manipulation, e.g. dragging, rotation, expansion or change of colour
• CPC Class Codes: G06F 3/0488 using a touch-screen or digitiser, e.g. input of commands through traced gestures
• CPC Class Codes: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images
• CPC Class Codes: G06V 30/412 Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables
• CPC Class Codes: G06V 30/414 Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text

Abstract

Systems and methods include an application operating on a device. The application causes the graphic user interface of the device to display an initial instruction to obtain a full-view image that positions all of an item within a field of view of a camera on the device. The application automatically recognizes identified features of the full-view image, by using a processor in communication with the camera. After displaying the initial instruction, the application causes the graphic user interface to display a subsequent instruction to obtain a zoom-in image that positions only a portion of the item within the field of view of the camera. Also the application automatically recognizes patterns from the zoom-in image, using the processor. Furthermore the application performs an authentication process using the identified features and the patterns to determine whether the item is valid, using the processor.

Claims

1. A method comprising:

causing a graphic user interface of a device to display an initial instruction to obtain continuous video that positions all of an item within a field of view of a camera of said device;

automatically recognizing identified features of said item from a full-view video frame of said continuous video obtained when all of said item was within said field of view of said camera, using a processor in communication with said camera;

after displaying said initial instruction, causing said graphic user interface to display a subsequent instruction to zoom in on said item and position only a portion of said item within said field of view of said camera while continuing to obtain said continuous video;

automatically recognizing patterns from a zoom-in video frame of said continuous video obtained when only said portion of said item occupied all of said field of view of said camera, using said processor; and

performing an authentication process using said identified features and said patterns to determine whether said item is valid, using said processor.

2. The method according to claim 1, further comprising determining whether said zoom-in video frame is actually of said item based on whether said continuous video is unbroken between said full-view frame and said zoom-in frame.

3. The method according to claim 1, further comprising:

determining a classification of said item based on said identified features; and

determining whether said item is valid based on said classification of said item matching a valid classification.

4. The method according to claim 1, further comprising:

monitoring said continuous video for continuity while said continuing to obtain said continuous video to determine said continuous video being discontinuous between said full-view frame and said zoom-in frame; and

after displaying said subsequent instruction and based on said continuous video being discontinuous, causing said graphic user interface to repeat said initial instruction, and subsequently repeat said subsequent instruction to zoom in.

5. The method according to claim 1, said full-view video frame being of insufficient quality to perform said recognizing patterns.

6. A method comprising:

causing a graphic user interface of a device to display an initial instruction to obtain a full-view image that positions all of an item within a field of view of a camera of said device;

automatically recognizing identified features of said full-view image, using a processor in communication with said camera;

after displaying said initial instruction, causing said graphic user interface to display a subsequent instruction to obtain a zoom-in image that positions only a portion of said item within said field of view of said camera;

automatically recognizing patterns from said zoom-in image, using said processor; and

performing an authentication process using said identified features and said patterns to determine whether said item is valid, using said processor.

7. The method according to claim 6, further comprising determining whether said zoom-in image is actually of said item based on an overlap of image features between said full-view image and said zoom-in image.

8. The method according to claim 6, further comprising:

determining a classification of said item based on said identified features; and

determining whether said item is valid based on said classification of said item matching a valid classification.

9. The method according to claim 6, further comprising:

monitoring said zoom-in image for continuity with said full-view image based on an overlap of image features between said full-view image and said zoom-in image to determine said zoom-in image being discontinuous with said full-view image; and

after displaying said subsequent instruction and based on said zoom-in image being discontinuous with said full-view image, causing said graphic user interface to repeat said initial instruction to obtain said full-view image, and subsequently repeat said subsequent instruction to obtain said zoom-in image.

10. The method according to claim 6, said full-view image being of insufficient quality to perform said recognizing patterns.

11. A system comprising:

an application operating on a device,

said application causing a graphic user interface of said device to display an initial instruction to obtain continuous video that positions all of an item within a field of view of a camera of said device,

said application automatically recognizing identified features of said item from a full-view video frame of said continuous video obtained when all of said item was within said field of view of said camera, using a processor in communication with said camera,

after displaying said initial instruction, said application causing said graphic user interface to display a subsequent instruction to zoom in on said item and position only a portion of said item within said field of view of said camera while continuing to obtain said continuous video,

said application automatically recognizing patterns from a zoom-in video frame of said continuous video obtained when only said portion of said item occupied said field of view of said camera, using said processor, and

said application performing an authentication process using said identified features and said patterns to determine whether said item is valid, using said processor.

12. The system according to claim 11, said application determining whether said zoom-in video frame is actually of said item based on whether said continuous video is unbroken between said full-view frame and said zoom-in frame.

13. The system according to claim 11, said application determining a classification of said item based on said identified features, and determining whether said item is valid based on said classification of said item matching a valid classification.

14. The system according to claim 11, said application:

monitoring said continuous video for continuity while said continuing to obtain said continuous video to determine said continuous video being discontinuous between said full-view frame and said zoom-in frame, and

after displaying said subsequent instruction and based on said continuous video being discontinuous, causing said graphic user interface to repeat said initial instruction, and subsequently repeat said subsequent instruction to zoom in.

15. The system according to claim 11, said full-view video frame being of insufficient quality to perform said recognizing patterns.

16. A system comprising:

an application operating on a device,

said application causing a graphic user interface of a device to display an initial instruction to obtain a full-view image that positions all of an item within a field of view of a camera of said device,

said application automatically recognizing identified features of said full-view image, using a processor in communication with said camera,

after displaying said initial instruction, said application causing said graphic user interface to display a subsequent instruction to obtain a zoom-in image that positions only a portion of said item within said field of view of said camera,

said application automatically recognizing patterns from said zoom-in image, using said processor, and

said application performing an authentication process using said identified features and said patterns to determine whether said item is valid, using said processor.

17. The system according to claim 16, said application determining whether said zoom-in image is actually of said item based on an overlap of image features between said full-view image and said zoom-in image.

18. The system according to claim 16, said application determining a classification of said item based on said identified features, and determining whether said item is valid based on said classification of said item matching a valid classification.

19. The system according to claim 16, said application:

monitoring said zoom-in image for continuity with said full-view image based on an overlap of image features between said full-view image and said zoom-in image to determine said zoom-in image being discontinuous with said full-view image, and

after displaying said subsequent instruction and based on said zoom-in image being discontinuous with said full-view image, causing said graphic user interface to repeat said initial instruction to obtain said full-view image, and subsequently repeat said subsequent instruction to obtain said zoom-in image.

20. The system according to claim 16, said full-view image being of insufficient quality to perform said recognizing patterns.