Histogram methods and systems for object recognition

• US 8,004,576 B2
• Filed: 05/19/2009
• Issued: 08/23/2011
• Est. Priority Date: 10/31/2008
• Status: Active Grant

• Current Assignee: Digimarc Corporation
• Original Assignee: Digimarc Corporation
• Inventors: Sharma, Ravi K.
• Primary Examiner(s): WHIPKEY, JASON T
• Application Number:
  US12/468,402
• Publication Number:
  US 20100113091A1
• Time in Patent Office: 826 Days
• Field of Search: 348/222.1, 348/211.2, 382/165, 382/168, 382/170
• US Class Current: 348/222.1
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: H04M 2250/52 including functional features of a camera constructional features concerning the integration of cameras in portable phones H04M1/0264

Abstract

A multidimensional histogram is used to characterize an image (or object), and is used to identify candidate matches with one or more reference images (or objects). An exemplary implementation employs hue information for two of the dimensions, and a second derivative function based on luminance for a third dimension. The simplicity and speed of the detailed arrangements make them well suited for use with cell phones and other mobile devices, which can use the technology for image/object recognition, e.g., in visual search applications.

Claims

1. A cell phone comprising:

a processor;

a memory;

a camera system; and

a wireless interface;

the memory storing instructions causing the processor to perform the following operations;

obtain a frame of image data from the camera system;

generate multi-dimensional histogram data from the obtained frame of image data, the histogram data having plural dimensions and defining a greater plurality of histogram bins, first and second dimensions of the plural dimensions corresponding to quantized hue information, a third dimension corresponding to quantized illumination information processed with a second derivative function;

compare the generated histogram data with plural sets of reference histogram data; and

by reference to the comparison, identify one or more candidate matching images corresponding to the captured image.

2. The cell phone of claim 1 in which the quantized illumination information is quantized into three levels.

3. The cell phone of claim 1 in which the quantized illumination information is quantized into two levels.

4. The cell phone of claim 1 in which the instructions stored in the memory cause the processor to normalize the generated histogram data to a reference, and then compute a histogram intersection measure between the normalized histogram data and each plural sets of reference histogram data.

5. The cell phone of claim 1 in which the instructions stored in the memory cause the processor to normalize the generated histogram data to a reference, and then compute a histogram match measure between the normalized histogram data and each plural sets of reference histogram data.

6. A system comprising:

a cell phone, including a processor, a memory, a camera system, and a wireless interface; and

a remote computer, including a processor, a memory, a camera system, and an interface permitting exchange of data with the cell phone;

the memories storing instructions causing the processors to perform the following operations;

obtain a frame of image data from the camera system;

generate multi-dimensional histogram data from the obtained frame of image data, the histogram data having plural dimensions and defining a greater plurality of histogram bins, first and second dimensions of the plural dimensions corresponding to quantized hue information, a third dimension corresponding to quantized illumination information processed with a second derivative function;

compare the generated histogram data with plural sets of reference histogram data; and

by reference to the comparison, identify one or more candidate matching images corresponding to the captured image.