Image analysis – Histogram processing – With a gray-level transformation
Reexamination Certificate
2001-02-26
2004-11-16
Au, Amelia M (Department: 2621)
Image analysis
Histogram processing
With a gray-level transformation
C382S132000, C382S274000
Reexamination Certificate
active
06819794
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an image processing apparatus, an image processing system, an image processing method, and a storage medium computer-readably storing processing steps for implementing the method that are used in, for example, an apparatus or a system for capturing a medical digital X-ray image.
2. Related Background Art
With digitization of images in the recent years, progress has been made in digitization of, for example, a medical X-ray image, and it has become possible to obtain a spatial distribution of X-ray intensity as a digital image (digital X-ray image).
As a method for capturing an X-ray digital image, there is such a method as a method for capturing an X-ray digital image through an exciting light distribution by a laser by forming a latent image with respect to an accelerated phosphorescence by X-ray energy, a method of converting a spatial distribution of X-ray intensity to an optical intensity distribution (phosphorous) and directly to an electric signal by an image sensor having a plurality of pixels, and then to an X-ray digital image, or a method for capturing an X-ray digital image after converting a special distribution of X-ray intensity directly to a distribution of electric charge.
There are following and various other advantages of capturing an X-ray digital image (digitizing an X-ray image):
storage and transfer of image data can be efficient;
an optimum image can be easily created by digital image processing (thus, failure in photographing can be easily recovered);
medical image diagnosis can be highly efficient; and
costs of medical image diagnosis can be reduced.
In particular, the advantage in that an optimum image can be easily created by digital image processing is the most important advantage of using an X-ray digital image in the field of the medical image diagnosis, and diagnosis by an X-ray digital image is impossible without this digital image processing.
More specifically, in a conventional method of analogically outputting an X-ray image onto a film, a clear and high-contrast image is created utilizing a field with sensitivity exposure density characteristic with respect to X-ray intensity (a field with high gamma) of a film. However, control for this purpose is mainly performed by conditions setting in X-ray photographing, when allowable ranges of photographing conditions are narrow.
For example,
FIG. 8
shows a relation “C” between a numerically indicated incident X-ray dose intensity distribution and film density to be exposed and developed when the X-ray is incident with an X-ray dose in the horizontal axis and film density in the vertical axis.
Here, in
FIG. 8
, if an optimum X-ray dose intensity distribution such as the one shown by “A” is obtained, an optimum image having a density distribution as shown by “B” that makes it easier for an observer (a physician or the like) to view. However, when photographing is performed with an inappropriate X-ray dose (such as high bulb voltage leading to short wavelength (hard line), and to less human body absorption), a dynamic range of the X-ray dose intensity distribution is narrowed as shown by “D”, and an image with an appropriate gradation cannot be obtained. Such a phenomenon can occur even if a thickness of a human body being a subject is thin. In addition, as shown by “E”, if an X-ray dose is appropriate but an exposure does is excessive, the entire X-ray dose intensity distribution shifts, and an image with an appropriate density distribution cannot be obtained in this case as well.
On the other hand, in an X-ray digital image, various kinds of X-ray dose intensity distributions as shown in
FIG. 9
can be obtained once as a digital value. Then, an image having an optimum gradation characteristic (a density characteristic on a film if hard-copied) as indicated by “B” in
FIG. 9
can be obtained by various kinds of gradation conversion characteristics (reference table) according to the X-ray intensity distribution as shown in “C
1
” to “C
3
” of
FIG. 2
, and the above-mentioned allowable range of photographing conditions can be substantially extended.
FIG. 7
shows an X-ray photographing apparatus
800
for capturing an X-ray digital image having the above-mentioned advantages.
In the X-ray photographing apparatus
800
, in the case in which a human body
802
lying on a table is photographed as a subject, an X-ray sensor panel
803
converts X-ray dose intensity distributions from an X-ray bulb
801
penetrating through the human body
802
to electric charge distributions and sequentially outputs them. An analog/digital converter
805
digitizes the output of the X-ray sensor panel
803
, and stores its digital image data (X-ray digital image data) in a memory
806
once. At this point, a controller
804
controls the timing of exposure by the X-ray bulb
801
and capturing an image.
Here, there are dispersions of an offset and a gain for each pixel in the X-ray sensor panel
803
. In order to correct the dispersions, an offset value being an image captured without irradiating an X-ray by the X-ray bulb
801
is stored in the memory
808
. In addition, a logarithmically converted version of a gain value being an image captured in the state in which there is not subject (human body
802
) is stored in a memory
809
.
A conversion unit
807
is for performing logarithmic conversion, and more specifically is a reference table (look-up table).
Therefore, the X-ray digital image data once stored in the memory
806
is logarithmically converted by the conversion unit
807
after the offset in the memory
808
is subtracted by a subtractor
811
, and becomes an X-ray intensity distribution image when a difference (division) between the data and the gain of the memory
809
is found by a subtractor
812
. This X-ray intensity distribution image is stored in a memory
810
once. Thereafter, the image data stored in the memory
810
is read out from storage, processing, display, hard copy or the like of an image, and is used for medical image diagnosis or the like.
At this point, gradation conversion processing in accordance with a gradation conversion characteristic shown in
FIG. 9
is applied to the image data read out from the memory
810
, and a gradation conversion characteristic is determined as follows in this gradation conversion processing according to a state at the time of obtaining the image data (image data).
(1) The gradation conversion characteristic is determined such that a density value (output pixel value) of a single or a plurality of arbitrary portions to be designated in an object image is an aimed value.
(2) A histogram of the object image is analyzed to extract a characteristic point of the histogram, and the gradation conversion characteristic is determined such that the characteristic point is an aimed value.
In these methods (1) and (2), a gradation conversion characteristic function is represented by a function having a small number of parameters, and the parameters are determined such that the parameters has a characteristic closest to an aimed value after subtracting a degree of freedom.
However, the conventional method such as the above-mentioned (1) or (2) has unstableness of gradation conversion processing because a complicated operation such as analysis of an object image itself analysis of a histogram of an object image intervenes or the like, a lot of time is required for analysis, computational processing or the like, and in some object images, an analysis mistake may occur.
In addition, for example, in the X-ray photographing apparatus (medical X-ray photographing apparatus) shown in
FIG. 10
, there is a photographing menu for efficiently performing photographing and image processing, and a user selects an item corresponding to a field of a subject (human body field) to be photographed from the photographing menu in advance. Further, there is a setting item for gradation conversion processing peculiar to each field (setting of a parameter to be used for the gradation conversion proc
Au Amelia M
Canon Kabushiki Kaisha
Fitzpatrick ,Cella, Harper & Scinto
LaRose Colin
LandOfFree
Image processing apparatus, image processing method, storage... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image processing apparatus, image processing method, storage..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing apparatus, image processing method, storage... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3354056