Image analysis – Pattern recognition – Template matching
Reexamination Certificate
2000-02-17
2004-01-13
Au, Amelia M. (Department: 2623)
Image analysis
Pattern recognition
Template matching
C382S217000, C382S219000, C382S220000
Reexamination Certificate
active
06678414
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of Invention
This invention relates to systems and methods for processing images using filters. More specifically, this invention relates to systems and methods for designing and implementing image processing filters using templates wherein the filters operate on gray-scale images and the templates identify gray-scale image features for the purposes of modification or for extracting some image statistic, and for the purposes of optimization for the human visual system, or compatibility with other system modules, such as, compression algorithms, recognition algorithms and those occurring in printing and display devices.
2. Description of Related Art
A wide variety of digital document processing tasks are performed using template-based filters. Illustratively, digital document processing tasks include resolution conversion, enhancement, restoration, appearance tuning and de-screening of images. These tasks are commonly performed on monochrome and color images, as well as binary and continuous tone images. Although, due to the binary nature of conventional templates, implementing many digital document-processing tasks on continuous tone images has been problematic prior to the present invention. A continuous tone image may also be referred to as a gray-scale image.
In conventional systems and methods, a typical filter includes template operators to perform filtering of the images, where, a filter may be characterized as an operator or device that transforms one image into another image or transforms an image to a collection of information, such as image statistics. The filter is formed of a number of imaging template operators, often simply referred to as templates. These templates may be, for example, stored in a look-up table and implemented using a look-up table formalism. Or other equivalent formalisms, such as Boolean logic may be employed. The number of templates in a filter may vary between a small number of templates to thousands of templates. Due to its versatility in design, a look-up table is typically used to implement a template-based filter.
A raster is a one-dimensional array of image data, reflecting a single line of data across a single dimension, i.e., the length or the width, of the image. Further, each location, or “picture element,” in an image may be called a “pixel.” In an array defining an image in which each item of data provides a value, each value indicating the properties of a location may be called a pixel value. Each pixel value is a bit in a binary form of an image, a gray-scale value in a gray-scale form of an image, or a set of color-spaced coordinates in a color coordinate form of an image. The binary form, gray-scale form, and color coordinate form are each arranged typically in a two-dimensional array, which defines an image. An N-dimensional array is typically used for an N-dimensional images, where for example, N=3 for 3-dimensional topographic images.
Using the typical binary image processing setting as an example, the filter, using the templates, transforms certain observed pixel patterns in a binary image, for example, into a corresponding enhanced binary pixel pattern. Specifically, the filter observes an arrangement of pixels using a suitable window or mask. A window is an imaging algorithmic device that observes a plurality of pixels at the same time, where the plurality of pixels is located about a target pixel. The values and locations of the observed pixels are inputted into the template matching operations. After observing the arrangement of pixels, about a target pixel, the filter then attempts to match the observed pixel pattern with one or more of the templates in the look-up table. If the look-up table contains a match to the observed pixel pattern, the look-up table generates an appropriate output. The output may be an enhanced pixel pattern for the target pixel that corresponds to the observed pixel pattern. The output could also be information in other forms; for example, the output could be a code denoting the match condition, or a data to be used for a statistical characterization of image regions.
A wide variety of types and sizes of observation windows or masks are known. The particular window used in a particular application depends on the image to be analyzed and the particular process to be performed on the image. Illustratively, a 3×3 window may be used to process an image. The 3×3 window, at various locations in the image, observes a 3×3 block, i.e., a 9-pixel block, of binary-valued pixels, for example. One pixel in the window is the target pixel, which is typically the center pixel, while the other pixels in the window are the neighboring pixels. The target pixel and the neighboring pixels form a neighborhood. The window is typically scanned across an image advancing from target pixel to target pixel.
After the neighborhood is observed in the window, the neighborhood is then processed in some manner. For example, the observed neighborhood may be transformed into a vector. The vector is expressed in the form of (x
1
, x
2
. . . x
N
) where N As the number of pixels in the neighborhood and is used to represent the properties of the target pixel, including the neighborhood of the target pixel. Each element of the vector represents one of the pixels observed in the window. The vector is then used in the look-up table to generate a desired output, for example.
A look-up table may be created in a wide variety of ways. Typically, an input value is input into the look-up table and, in response, the look-up table outputs an output value. Further, the look-up table is typically created using a training image or a set of training images. “Restoration and Enhancement of Digital Documents,” by R. Loce and E. Dougherty, teaches a variety of methods for designing templates based on sets of training images. The training images will occur in pairs, where one member is the “typically input image,” or the “typically observed image,” i.e., the “observed image,” and the other image is the “ideal desired processed version of the image,” i.e., the “ideal image.” The training image pairs may be input into a computer program that acquires and analyzes pattern statistics between the two images, i.e., using computer-aided filter design techniques.
Conventional computer-aided filter design may be accomplished through using training-sets of document bitmaps, for example.
Illustratively, for designing a filter that enhances from a binary state to a gray-scale state, for a given pattern that occurs in the binary image about a target pixel, a training analysis system examines a target pixel at that corresponding location in the gray-scale image. The center of the window may be placed at the target pixel, for example. Based on the set of gray-scale pixels in the gray-scale image that are associated with corresponding target pixels in the binary image and gray-scale image, and associated with a similar neighborhood pixel pattern, a “best gray-scale pixel value” is determined for processing a target pixel that possess that pattern of neighborhood pixels. In other words, a template is created for the target pixels in the binary image possessing similar neighborhood pixel patterns. This analysis is performed for all binary patterns that are significant.
In this process of template selection, significance may be due to attributes such as the pixel pattern's frequency of occurrence, the pattern's effect on the generated image, or both. Accordingly, if a template, i.e., a pattern of pixels in the binary image, is considered significant with respect to template inclusion in the design process, that template will appear in the template-matching filter. Upon operating on an input image, if that pattern is observed, the observed target pixel value will be assigned or associated with a certain value, i.e., a corresponding gray-scale value. Both the observed neighborhood and the corresponding gray-scale value may be stored in the look-up table. Accordingly, the look-up table accepts inpu
Cuciurean-Zapan Clara
Loce Robert P.
Williams Leon C.
Au Amelia M.
Blair Philip E.
Krishnan Aditya
Le Brian Q.
Xerox Corporation
LandOfFree
Loose-gray-scale template matching does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Loose-gray-scale template matching, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Loose-gray-scale template matching will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3191202