US5633511A — Automatic tone scale adjustment using image activ…

Claim 1

Independent Claim 110 words

A method of automatic tone scale adjustment for digital radiographic images comprising the steps of: providing an input digital radiographic image; estimating the image noise as a function of code value of said image; determining the range of code values in said input image; determining which way said input image should be processed; determining a bounding box which contains the region of interest of said input image; computing the image activity histogram of said input image; determining four points from said image activity histogram; constructing the tone-scale curve for said input image; and mapping said input image through said tone-scale curve to produce an output image with good tone scale.

Claim 2

Independent Claim 110 words

A method of automatic tone scale adjustment for digital radiographic images comprising the steps of: providing an input digital radiographic image; estimating the image noise as a function of code value of said image; determining the range of code values in said input image; determining which way said input image should be processed; determining a bounding box which contains the region of interest of said input image; computing the image activity histogram of said input image; determining four points from said image activity histogram; constructing the tone-scale curve for said input image; and mapping said input image through said tone-scale curve to produce an output image with good tone scale.

Claim 3

Dependent Claim 25 words

The method of claim 1 wherein said providing step includes converting an x-ray image stored in a storage phosphor into said input digital radiographic image.

Claim 4

Dependent Claim 24 words

The method of claim 1 wherein said providing step includes digitizing a radiographic image on x-ray film to produce said input digital radiographic image.

Claim 5

Dependent Claim 46 words

The method of claim 1 wherein said estimating step includes the steps of exposing a gray scale of uniform patches at different exposures, calculating noise from the standard deviation of the code values in each uniformly exposed area, and interpolating the noise measurements of neighboring exposures.

Claim 6

Dependent Claim 65 words

The method of claim 1 wherein said estimating step includes the steps of computing the edge gradient of every pixel of the input image to distinguish the uniform area from the busy area in said image, estimating the noise value from the standard deviation in the uniform area, and interpolating the noise at any level of code value from the noise measurement of neighboring exposures.

Claim 7

Dependent Claim 97 words

The method of claim 1 wherein said step of determining the range of code values in said input image includes the steps of: calculating the code-value histogram, setting mincode=the smallest code value in the histogram that has more than 10 pixels, setting maxcode=the largest code value in the histogram that has more than 10 pixels, finding the background peak within an adaptive interval below maxcode, if the background peak is found, determining the first valley to the left of the peak as the right-bound, if the peak is not found, then setting right-bound=maxcode-2* NOISE, setting the left-bound=mincode+2*NOISE.

Claim 8

Dependent Claim 89 words

The method of claim 1 wherein said step of determining which way said input image should be processed includes the steps of: extracting the most modulated segments from all image lines in the horizontal direction and calculate the average length, extract the most modulated segments from all image lines in the vertical direction and calculate the average length, and if the average length of the horizontal segments is shorter than that of the vertical segments, then the image activity will be calculated with the horizontal segments, or else, vertical.

Claim 9

Dependent Claim 103 words

The method of claim 7 wherein said extracting steps include the steps of: setting an adaptive threshold, CLIFF; partitioning an image line into intervals by a finite state machine; if, within an interval, the code value changes by an amount greater than CLIFF, then labeling the interval as a transition interval, and grouping other pixels on the image line into segments separated by the transition intervals and the image boundaries; fitting each segment with a linear function and using the residual error of the fit as a measure of signal modulation of that segment; and extracting the segment with the largest signal modulation.

Claim 10

Dependent Claim 111 words

The method of claim 1 wherein said step of determining the bounding box includes the steps of: extracting the most modulated segments from all image lines; determining the spatial spread of all the pixels in the extracted image segments, the mean and standard deviation along the x- and the y- directions; determining the width and height of the rectangular bounding box, wherein width is 3 standard deviations in the x-direction and height is 3 standard deviations in the y-direction; and centering the bounding box at the x-mean and the y-mean, with the width and height being adjusted so that they are within preselected ranges and also well within the image boundary.

Claim 11

Dependent Claim 63 words

The method of claim 1 wherein said step of computing the image activity histogram includes the steps of: processing the digital image on a line by line basis; calculating image activity at each pixel for each image line; and accumulating the image activity for each code value from each image line to form a cumulative activity histogram as a function of code value.

Claim 12

Dependent Claim 25 words

The method of claim 1 wherein said step of computing includes the step of measuring image activity by calculating the Laplacian contrast for each pixel.

Claim 13

Dependent Claim 25 words

The method of claim 1 wherein said step of computing includes the step of measuring image activity by calculating the edge gradient of each pixel.

Claim 14

Dependent Claim 23 words

The method of claim 1 wherein said step of computing includes the step of measuring image activity by calculating the code-value co-occurrence histogram.

Claim 15

Dependent Claim 70 words

The method of claim 1 wherein said step of computing includes the steps of measuring image activity by counting the level-crossings for each code value by, for each image line, updating an array that keeps track of the number of sign changes for each code value, and accumulating the number of sign changes from each image line to form a cumulative histogram of level-crossings as a function of code value.

Claim 16

Dependent Claim 97 words

The method of claim 1 wherein said step of determining the four points includes the steps of: within the range between left-bound and right-bound, finding the highest peak, p, in the activity histogram; normalizing the amplitude of the highest peak to 1.0; locating the rightmost peak within 1.0 log exposure above p and with normalized amplitude higher than 0.3; finding the right point (rp) and the far right point (frp); locating the leftmost peak within 1.0 log exposure below p and with normalized amplitude higher than 0.3; and finding the left point and the far-left point (flp).

Claim 17

Dependent Claim 108 words

The method of claim 1 wherein said step of constructing the tone scale curve includes the steps of; determining the brightness vs. Log E curve from the formula Brightness=a*logE+b by determining the slope a and the intercept b from the left point lp and the right point rp by mapping lp and rp to preselected densities; mapping brightness to luminance using the perceptual brightness model ##EQU1## mapping luminance to film density, where D=log Source-log Luminance; constructing the visually optimized tone scale curve, D=f(log E), that renders equal log exposure differences as equal brightness differences; and generating the toe and shoulder tails with the following equation D=(Dmax-Dmin)/(1+[E/E0] d )+Dmin.

Claim 18

Dependent Claim 54 words

The method of claim 1 wherein, in said determining four points step, left and right feature points are determined and including the step of producing a reversed-polarity tone scale by exchanging roles of said left and right feature points for a normal tone-scale algorithm and then reversing the polarity of the input code values.

Claim 19

Independent Claim 130 words

A method of extracting the important range of code values from an input digital image by computing image activity as a function of code value, comprising the steps of: (a) processing the digital image on a line by line basis; (b) calculating image activity at each pixel for each image line; (c) accumulating the image activity for each code value from each image line to form a cumulative activity histogram as a function of code value; and (d) locating the highest peak(s) of the activity histogram and one point on each side (the left and the right) of the peak(s) wherein the histogram height falls below certain thresholds; and (e) extracting the range of code values between the left point and the right point, as identified in the previous step.

Claim 20

Independent Claim 130 words

A method of extracting the important range of code values from an input digital image by computing image activity as a function of code value, comprising the steps of: (a) processing the digital image on a line by line basis; (b) calculating image activity at each pixel for each image line; (c) accumulating the image activity for each code value from each image line to form a cumulative activity histogram as a function of code value; and (d) locating the highest peak(s) of the activity histogram and one point on each side (the left and the right) of the peak(s) wherein the histogram height falls below certain thresholds; and (e) extracting the range of code values between the left point and the right point, as identified in the previous step.

Claim 21

Dependent Claim 33 words

The method of extracting the important range of code values from an input digital image claimed in claim 18, wherein the image activity is measured by calculating the Laplacian contrast for each pixel.

Claim 22

Dependent Claim 33 words

The method of extracting the important range of code values from an input digital image claimed in claim 18, wherein the image activity is measured by calculating the edge gradient for each pixel.

Claim 23

Dependent Claim 31 words

The method of extracting the important range of code values from an input digital image claimed in claim 18, wherein the image activity is measured by calculating the code-value co-occurrence histogram.

Claim 24

Independent Claim 143 words

A method of extracting the important range of code values from an input digital image by counting the number of level-crossings for each code value, comprising the steps of: (a) processing the digital image on a line by line basis; (b) for each image line, updating an array that keeps track of the number of sign changes for each code value; (c) accumulating the number of sign changes for each code value from each image line to form a cumulative histogram of level-crossings as a function of code value; and (d) locating the highest peak of the level-crossing histogram and one point on each side (the left and the right) of the peak where the histogram height falls below certain criteria; and (e) extracting the range of code values between the left point and the right point, as identified in the previous step.

Claim 25

Independent Claim 143 words

A method of extracting the important range of code values from an input digital image by counting the number of level-crossings for each code value, comprising the steps of: (a) processing the digital image on a line by line basis; (b) for each image line, updating an array that keeps track of the number of sign changes for each code value; (c) accumulating the number of sign changes for each code value from each image line to form a cumulative histogram of level-crossings as a function of code value; and (d) locating the highest peak of the level-crossing histogram and one point on each side (the left and the right) of the peak where the histogram height falls below certain criteria; and (e) extracting the range of code values between the left point and the right point, as identified in the previous step.

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	2000-05-26	2000-11-25	2001-05-27	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	2004-05-26	2004-11-25	2005-05-27	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2008-05-26	2008-11-25	2009-05-27	$7700.00	Window Closed / Lapsed

Automatic tone scale adjustment using image activity measures

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag