Methods and systems for digital image characteristic adjustment using a neural network

• US 7,221,807 B2
• Filed: 03/29/2002
• Issued: 05/22/2007
• Est. Priority Date: 03/29/2002
• Status: Expired due to Term

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Campbell, Richard John
• Primary Examiner(s): Bella; Matthew C.
• Assistant Examiner(s): PERUNGAVOOR, SATHYANARAYA V
• Application Number:
  US10/112,608
• Publication Number:
  US 20030185457A1
• Time in Patent Office: 1,880 Days
• Field of Search: 382155-161, 382/274
• US Class Current: 382/274
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/60 using machine learning, e.g. neural networks
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: H04N 1/4074 using histograms
• CPC Class Codes: H04N 1/6005 with luminance or chrominance signals, e.g. LC1C2, HSL or YUV

Abstract

Embodiments of the present invention comprise methods and systems for automatically adjusting images to conform to preference data.

Claims

1. A method for automatically calculating an image-specific correction curve for a subject image using a neural network, said method comprising:

receiving histogram data for a subject image;

inputting said histogram data to a neural network that has been trained by a user adjusting a tone enhancement curve by manipulating the shape of said tone enhancement curve through a user input apparatus to produce a preferred output;

generating a plurality of interpolation points for an image-specific correction curve with said neural network;

determining said image-specific correction curve by performing interpolation using said plurality of said interpolation points; and

outputting said image-specific correction curve.

2. A method as described in claim 1 further comprising generating a correction table from said image-specific correction curve.

3. A method as described in claim 2 further comprising applying said correction table to said subject image.

4. A method as described in claim 2 wherein said correction table is a 1-D Look-Up-Table (LUT).

5. A method as described in claim 1 further comprising generating said subject image histogram data.

6. A method as described in claim 1 wherein said neural network has been trained with observer preference data that is linked to image adjustment data through a database.

7. A method as described in claim 1 wherein said neural network comprises two neuron layers.

8. A method as described in claim 7 wherein said neural network comprises a tangent sigmoid transfer function in the first neuron layer.

9. A computer readable medium comprising computer-executable instructions for automatically calculating an image-specific correction curve for a subject image using a neural network, said method comprising:

receiving histogram data for a subject image;

inputting said histogram data to a neural network that has been trained by a user adjusting a tone enhancement curve by manipulating the shape of said tone enhancement curve through a user input apparatus to produce a preferred output;

generating a plurality of interpolation points for an image-specific correction curve with said neural network; and

determining said image-specific correction curve by performing interpolation using said plurality of said interpolation points.

10. A method for automatically calculating an image-specific correction table for a subject image using a neural network, said method comprising:

receiving a subject image;

converting said subject image into a luminance-segregated format;

generating a subject image luminance channel histogram;

inputting said histogram into a neural network that has been trained by a user adjusting a tone enhancement curve by manipulating the shape of said tone enhancement curve through a user input apparatus to produce a preferred output;

calculating a plurality of interpolation points that define an image-specific correction curve for said subject image with said neural network;

determining said image-specific correction curve by performing interpolation using said plurality of said interpolation points;

establishing a correction table from said image-specific correction curve; and

processing said subject image with said correction table.