Two-bit morphology processing apparatus and method

• US 6,038,352 A
• Filed: 02/19/1998
• Issued: 03/14/2000
• Est. Priority Date: 10/17/1997
• Status: Expired due to Term

• Current Assignee: Microscan Systems Inc (Omron Corporation)
• Original Assignee: Acuity Brands, Inc.
• Inventors: Wilt, Michael J.
• Primary Examiner(s): Au, Amelia
• Assistant Examiner(s): Wu, Jingge
• Application Number:
  US09/026,053
• Time in Patent Office: 754 Days
• Field of Search: 382/302, 382/303, 382/304, 382/307, 382/308, 382/282, 382/283
• US Class Current: 382/308
• CPC Class Codes: G06T 1/60 Memory management
• CPC Class Codes: G06T 5/30 Erosion or dilatation, e.g. thinning

Abstract

The present invention provides a novel system and method that permits a "mask" to be directly incorporated into an image during image processing. This is accomplished by processing binary images or image data which are encoded using two bits rather than the usual one. The second bit is defined to be a "mask enable", which directs a processor to pass the original data through to the output image regardless of the processing result for that pixel. The present invention also provides a means of automatically providing background data to the processor for pixels outside the original image so that the result image is always the same size as the original image. For binary images, the background may be defined to have a value of "0" or "1", and this value is provided to the processing engine in place of all of the pixels which lie outside the original image. For gray-scale images, the minimum or maximum possible value is provided to the processing engine in place of all of the pixels which lie outside of the original image, effectively eliminating these values from consideration when the minimum or maximum of the neighborhood pixels is computed. The determination of whether a pixel is outside of the original image is implemented using the framing signals provided by a data flow controller along with the image data.

Claims

1. An improved system for performing binary morphology computations on one or more adjacent image pixels comprising:

a data flow controller for providing image data in groups of adjacent image pixels, said image data including at least two bits per pixel, a first bit of which contains a binary image value and a second bit of which contains a mask protection bit;

a neighborhood generator, which receives said image data and said framing signals and which generates neighborhood data; and

a neighborhood processor, for receiving neighborhood pixel data corresponding to a neighborhood of pixels in an image being processed, wherein said neighborhood pixel data is represented using at least first and second bits for each of said pixels, and wherein said first bit is an original binary pixel value and said second bit is a mask protection bit for protecting said original binary pixel value, and for outputting a final binary result, wherein said final binary result is said original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original pixel value, and wherein said final binary result is a result value from a binary function look-up table if said mask protection bit for said neighborhood of pixels is not set to protect said original pixel value.

2. A method for performing morphology computations on one or more adjacent pixels, said method comprising the steps of:

providing, from a data flow controller, image data in groups of adjacent pixels and framing signals including start-of-row, end-of-row and end-of-frame signals to a neighborhood generator, said image data including at least two bits per pixel, at least one bit of which contains a binary image value and at least one bit of which contains a mask bit;

receiving, by said neighborhood generator, said image data and said framing signals and generating neighborhood data signals; and

interpreting said image data and replacing the result of a processing computation with a copy of said binary image value bit if said mask bit so dictates, regardless of the result of said processing computation, said step of interpreting said image data comprising;

receiving said neighborhood data including said binary image value and said mask bit for each pixel in a neighborhood of pixels;

forming an integer index value corresponding to said neighborhood of pixels by concatenating said neighborhood pixel data;

retrieving a result value from a binary function look-up table based upon said integer index value; and

outputting a final binary result, wherein said final binary result is an original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original binary pixel value, and wherein said final binary result is said result value from said binary function look-up table if said mask protection bit for said neighborhood of pixels is not set to protect said original binary pixel value.

3. The method of claim 2 wherein said neighborhood generator generates neighborhood data and corresponding neighborhood pixel-valid signals.

4. The method of claim 2 wherein said step of generating neighborhood data and corresponding neighborhood pixel-valid signal comprises recycling said image data through at least one two-row FIFO (First-In, First-Out) memory.

5. A system for performing binary morphology processing using a mask protection bit, said system comprising:

a neighborhood processor, for receiving neighborhood pixel data corresponding to a neighborhood of pixels in an image being processed, wherein said neighborhood pixel data is represented using at least first and second bits for each of said pixels, and wherein said first bit is an original binary pixel value and said second bit is a mask protection bit for protecting said original binary pixel value, and for outputting a final binary result, wherein said final binary result is said original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original pixel value, and wherein said final binary result is a result value if said mask protection bit for said neighborhood of pixels is not set to protect said original pixel value.

6. The system of claim 5 further including a neighborhood generator, which receives said image pixel data and framing signals and which generates neighborhood data and corresponding neighborhood pixel-valid signals.

7. The system of claim 6 wherein said neighborhood generator comprises at least two FIFO (First In, First Out) memories through which said image pixel data is cycled.

8. An improved system for performing binary morphology computations on one or more adjacent image pixels comprising:

a data flow controller for providing image data in groups of adjacent image pixels, said image data including at least two bits per pixel, a first bit of which contains a binary image value and a second bit of which contains a mask bit, said data flow controller for also providing framing signals;

a neighborhood generator, comprising at least two FIFO (First-In, First-Out) memories, which receives said image data and framing signals and which generates neighborhood data and corresponding neighborhood pixel-valid signals; and

a neighborhood processor, for receiving neighborhood pixel data corresponding to a neighborhood of pixels in an image being processed, wherein said neighborhood pixel data is represented using at least first and second bits for each of said pixels, and wherein said first bit is an original binary pixel value and said second bit is a mask protection bit for protecting said original binary pixel value, and for outputting a final binary result, wherein said final binary result is said original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original pixel value, and wherein said final binary result is a result value from a binary function look-up table if said mask protection bit for said neighborhood of pixels is not set to protect said original pixel value.

9. The system of claim 8 wherein said at least one binary neighborhood processor comprises:

means for receiving said neighborhood data including said binary image value and said mask bit for each pixel in a neighborhood of pixels;

means for forming an integer index value corresponding to said neighborhood of pixels by concatenating said neighborhood pixel data;

means for retrieving a result value from a binary function look-up table based upon said integer index value; and

means for outputting a final binary result, wherein said final binary result is an original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original binary pixel value, and wherein said final binary result is said result value from said binary function look-up table if said mask protection bit for said neighborhood of pixels is not set to protect said original binary pixel value.

10. A binary neighborhood processor for performing binary morphology processing on neighborhood pixel data using a mask protection bit, said binary neighborhood processor comprising:

means for receiving said neighborhood data including an original binary pixel value and said mask protection bit for each pixel in a neighborhood of pixels;

means for forming an integer index value corresponding to said neighborhood of pixels by concatenating said neighborhood pixel data;

means for retrieving a result value from a binary function look-up table based upon said integer index value; and

means for outputting a final binary result, wherein said final binary result is an original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original binary pixel value, and wherein said final binary result is said result value if said mask protection bit for said neighborhood of pixels is not set to protect said original binary pixel value.

11. A method of performing binary morphology processing using an embedded mask, said method comprising the steps of:

receiving neighborhood pixel data corresponding to a neighborhood of pixels in an image being processed, wherein said neighborhood pixel data is represented using at least first and second bits for each of said pixels, and wherein said first bit is a binary pixel value and said second bit is a mask protection bit for protecting said binary pixel value;

forming an integer index value corresponding to said neighborhood of pixels by concatenating said neighborhood pixel data;

retrieving a result value from a binary function look-up table based upon said integer index value; and

outputting a final binary result, wherein said final binary result is an original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original pixel value, and wherein said final binary result is said result value from said binary function look-up table if said mask protection bit for said neighborhood of pixels is not set to protect said original pixel value.

12. A method of performing binary morphology processing using an embedded mask, said method comprising the steps of:

receiving neighborhood pixel data corresponding to a neighborhood of pixels in an image being processed, wherein said neighborhood pixel data is represented using at least first and second bits for each of said pixels, and wherein said first bit is a binary pixel value and said second bit is a mask protection bit for protecting said binary pixel value;

computing a result value based on said neighborhood pixel data; and

outputting a final binary result, wherein said final binary result is an original binary pixel value for said neighborhood of pixels if said mask protection bit for said neighborhood of pixels is set to protect said original pixel value, and wherein said final binary result is said result value if said mask protection bit for said neighborhood of pixels is not set to protect said original pixel value.