JP2000030044A - Three-dimensional image processor - Google Patents

Three-dimensional image processor

Info

Publication number
JP2000030044A
JP2000030044A JP10200754A JP20075498A JP2000030044A JP 2000030044 A JP2000030044 A JP 2000030044A JP 10200754 A JP10200754 A JP 10200754A JP 20075498 A JP20075498 A JP 20075498A JP 2000030044 A JP2000030044 A JP 2000030044A
Authority
JP
Japan
Prior art keywords
component
image
dimensional image
detailed
components
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP10200754A
Other languages
Japanese (ja)
Inventor
Shinichi Utsunomiya
眞一 宇都宮
Original Assignee
Shimadzu Corp
株式会社島津製作所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shimadzu Corp, 株式会社島津製作所 filed Critical Shimadzu Corp
Priority to JP10200754A priority Critical patent/JP2000030044A/en
Publication of JP2000030044A publication Critical patent/JP2000030044A/en
Pending legal-status Critical Current

Links

Abstract

PROBLEM TO BE SOLVED: To provide a three-dimensional image processor improving the quality of a three-dimensional image. SOLUTION: A multiple resolution breaking down part 11 executes wavelet transformation to each image signal of the X, Y and Z directions of the three- dimensional image stored in an original image memory 2 and breaks down the three-dimensional image into the approximate component of the three- dimensional image being a multiple resolution component and plural detailed components. A detailed component intensity transforming part 13 makes a previously set intensity transformation function to operate to an optional detailed component among plural detailed components to transform the intensity of the optional detailed component. An image reconstituting part 14 reconstitutes the three-dimensional image from plural detailed components including the intensely transformed detailed component and the approximate component. As the result, noise components included in the three-dimensional image can be reduced in a three-dimensional state and a prescribed component can be emphasized in the three-dimensional state, thereby the image quality is improved as the whole three-dimensional image.

Description

【発明の詳細な説明】 DETAILED DESCRIPTION OF THE INVENTION

【0001】 [0001]

【発明の属する技術分野】この発明は、X線断層撮影装
置や、磁気共鳴断層撮影装置などで被撮像対象を撮像す
ることで得られる複数の断層画像に基づく3次元画像に
画像処理を施す3次元画像処理装置に係り、特に、3次
元画像全体としての画質を向上させるための技術に関す
る。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for performing image processing on a three-dimensional image based on a plurality of tomographic images obtained by imaging an object to be imaged with an X-ray tomography apparatus, a magnetic resonance tomography apparatus, or the like. The present invention relates to a three-dimensional image processing apparatus, and more particularly to a technique for improving the image quality of a three-dimensional image as a whole.

【0002】[0002]

【従来の技術】X線断層撮影装置(X線CT)や、磁気
共鳴断層撮影装置(MRI)などの断層画像撮影装置で
被撮像対象の複数箇所を撮像することで2次元情報であ
る複数の断層画像を得る。これら複数の断層画像を積み
重ねることで、3次元構造の画像情報である3次元画像
が得られる。
2. Description of the Related Art A plurality of tomographic images, such as an X-ray tomography apparatus (X-ray CT) and a magnetic resonance tomography apparatus (MRI), are imaged at a plurality of locations on an object to be imaged to obtain a plurality of two-dimensional information. Obtain a tomographic image. By stacking the plurality of tomographic images, a three-dimensional image that is image information of a three-dimensional structure can be obtained.

【0003】従来、このような3次元画像の画質の改善
を行なう場合には、まず、上述した2次元情報である各
断層画像に対して、ラプラシアンフィルタなどを用いる
エッジ強調や、メディアンフィルタなどを用いる雑音低
減などの画像処理を施して、断層画像の画質を改善す
る。これら画質が改善された複数の断層画像に基づいた
3次元画像を得ることで、3次元画像の画質の改善を行
なっている。
Conventionally, in order to improve the image quality of a three-dimensional image, first, an edge enhancement using a Laplacian filter or a median filter is applied to each tomographic image as the two-dimensional information. Image processing such as noise reduction is performed to improve the image quality of the tomographic image. By obtaining a three-dimensional image based on a plurality of tomographic images with improved image quality, the image quality of the three-dimensional image is improved.

【0004】[0004]

【発明が解決しようとする課題】しかしながら、このよ
うな構成を有する従来例の場合には、次のような問題が
ある。従来、3次元画像内の2次元情報である断層画像
に対してだけ画像処理を施しているので、他の1次元方
向の情報である各断層画像間にまたがる面情報や雑音情
報などが無視されている。その結果、3次元画像全体と
して、各断層画像ごとに不自然に強調されるという問題
がある。また、その3次元画像の任意断層画像を切り出
した場合に、任意断層画像内が部分的に強調され、つな
がりのない断層画像となるという問題もある。
However, the prior art having such a structure has the following problems. Conventionally, image processing is performed only on a tomographic image that is two-dimensional information in a three-dimensional image, so that surface information or noise information that extends between each tomographic image that is other one-dimensional information is ignored. ing. As a result, there is a problem that the entire three-dimensional image is unnaturally emphasized for each tomographic image. Further, when an arbitrary tomographic image of the three-dimensional image is cut out, there is also a problem that the inside of the arbitrary tomographic image is partially emphasized, resulting in a disconnected tomographic image.

【0005】この発明は、このような事情に鑑みてなされたものであって、3次元画像に含まれる全ての情報を利用することで、3次元画像全体としての画質を向上することができる3次元画像処理装置を提供することを目的とする。 The present invention has been made in view of such circumstances, and can improve the image quality of the entire three-dimensional image by using all information included in the three-dimensional image. It is an object to provide a two-dimensional image processing device.

【0006】 [0006]

【課題を解決するための手段】この発明は、このような
目的を達成するために、次のような構成をとる。すなわ
ち、この発明は、被撮像対象の複数箇所で断層画像を撮
像することで得られる3次元画像を記憶する画像記憶手
段と、前記画像記憶手段に記憶された3次元画像の画像
信号に画像処理を施す画像処理手段とを備える3次元画
像処理装置において、前記画像処理手段は、(a)前記
3次元画像の各次元方向の画像信号にウェーブレット変
換を施すことにより、前記3次元画像を多重解像度成分
に分解して、前記3次元画像の近似成分と複数の詳細成
分とを求める多重解像度分解手段と、(b)前記多重解
像度分解手段で求められる複数の詳細成分の中の任意の
詳細成分に、予め設定された強度変換関数を作用させる
ことで、前記任意の詳細成分に含まれる雑音成分を低減
するとともに、所定成分を強調する強度変換を行なう強
度変換手段と、(c)前記強度変換手段で強度変換され
た任意の詳細成分を含む複数の詳細成分と、前記近似成
分とで構成される強度変換後多重解像度成分にウェーブ
レット逆変換を施すことにより、3次元画像を再構成す
る画像再構成手段とを備えたことを特徴とするものであ
る。
The present invention has the following configuration to achieve the above object. That is, the present invention provides an image storage unit that stores a three-dimensional image obtained by capturing a tomographic image at a plurality of locations on an imaging target, and performs image processing on an image signal of the three-dimensional image stored in the image storage unit. And (c) performing a wavelet transform on an image signal in each dimension of the three-dimensional image to perform multi-resolution image processing on the three-dimensional image. A multi-resolution decomposing means for decomposing the three-dimensional image into an approximate component and a plurality of detailed components, and (b) an arbitrary detailed component among the plurality of detailed components obtained by the multi-resolution decomposing means. Intensity conversion means for applying a predetermined intensity conversion function to reduce a noise component included in the arbitrary detailed component and perform intensity conversion for emphasizing a predetermined component. (C) performing a three-dimensional inverse wavelet transform on the multiresolution component after the intensity conversion composed of a plurality of detailed components including an arbitrary detailed component that has been intensity-converted by the intensity conversion means and the approximate component. Image reconstructing means for reconstructing an image.

【0007】〔作用〕この発明の作用は次のとおりであ
る。画像記憶手段は、2次元情報である断層画像を被撮
像対象の複数箇所で撮像することで得られる3次元画像
を記憶する。多重解像度分解手段は、この3次元画像の
各次元方向の画像信号にウェーブレット変換を施す。こ
のウェーブレット変換によって、3次元画像の近似成分
と、複数の詳細成分とである多重解像度成分が求まる。
強度変換手段は、予め設定された強度変換関数を、多重解像度分解手段で求められた任意の詳細成分に作用させることで、任意の詳細成分の雑音成分を低減するとともに、所定成分を強調する。 The intensity conversion means acts on an arbitrary detailed component obtained by the multi-resolution decomposition means by applying a preset intensity conversion function to reduce the noise component of the arbitrary detailed component and emphasize the predetermined component. 画像再構成手段は、雑音成分が適言されるとともに、所定成分が強調された任意の詳細成分を含む複数の詳細成分と、近似成分とで構成される強度変換後多重解像度成分にウェーブレット逆変換を施して、3次元画像を再構成する。 The image reconstructing means is wavelet-reverse-converted into a multi-resolution component after intensity conversion, which is composed of a plurality of detailed components including an arbitrary detailed component in which a predetermined component is emphasized and an approximate component while the noise component is properly stated. To reconstruct the three-dimensional image. この再構成された3 This reconstructed 3
次元画像は、所定成分が強調され、雑音成分が低減されたことによって、3次元画像全体として画質が向上している。 In the dimensional image, a predetermined component is emphasized and a noise component is reduced, so that the image quality of the dimensional image as a whole is improved. [Operation] The operation of the present invention is as follows. The image storage unit stores a three-dimensional image obtained by imaging a tomographic image, which is two-dimensional information, at a plurality of locations on an imaging target. The multi-resolution decomposition means performs a wavelet transform on the image signal in each dimension of the three-dimensional image. By this wavelet transform, a multi-resolution component, which is an approximate component of a three-dimensional image and a plurality of detailed components, is obtained. [Operation] The operation of the present invention is as follows. The image storage unit stores a three-dimensional image obtained by imaging a tomographic image, which is two-dimensional information, at a plurality of locations on an imaging target. The multi- resolution decomposition means performs a wavelet transform on the image signal in each dimension of the three-dimensional image. By this wavelet transform, a multi-resolution component, which is an approximate component of a three-dimensional image and a plurality of detailed components, is obtained.
The intensity conversion unit applies a predetermined intensity conversion function to an arbitrary detailed component obtained by the multi-resolution decomposition unit, thereby reducing a noise component of the arbitrary detailed component and enhancing a predetermined component. The image reconstruction means performs an inverse wavelet transform into a multi-resolution component after the intensity conversion, which includes a plurality of detailed components including an arbitrary detailed component in which a noise component is appropriately specified and a predetermined component is emphasized, and an approximate component. To reconstruct a three-dimensional image. This reconstructed 3 The intensity conversion unit applies a predetermined intensity conversion function to an arbitrary detailed component obtained by the multi-resolution decomposition unit, thereby reducing a noise component of the arbitrary detailed component and enhancing a predetermined component. The image reconstruction means performs an inverse wavelet transform into a multi-resolution component after the intensity conversion, which includes a plurality of detailed components including an arbitrary detailed component in which a noise component is appropriately specified and a predetermined component is emphasized, and an approximate component. To reconstruct a three-dimensional image. This reconstructed 3
In the three-dimensional image, the predetermined component is emphasized and the noise component is reduced, so that the image quality of the entire three-dimensional image is improved. In the three-dimensional image, the predetermined component is emphasized and the noise component is reduced, so that the image quality of the entire three-dimensional image is improved.

【0008】 [0008]

【発明の実施の形態】以下、図面を参照してこの発明の一実施例を説明する。図1はこの発明の3次元画像処理装置の概略構成を示すブロック図である。 DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of a three-dimensional image processing apparatus according to the present invention.

【0009】図1に示すように、本実施例の3次元画像
処理装置は、図示しない磁気共鳴断層撮影装置等で得ら
れる3次元画像を記憶する原画像メモリ2と、3次元画
像の各次元方向の画像信号に画像処理を施す画像処理部
1と、画像処理部1で利用される強度変換関数を入力す
る入力部3と、入力部3で入力された強度変換関数を記
憶する強度変換関数メモリ4と、画像処理部1で画像処
理された3次元画像を表示する表示部5とを備えてい
る。
As shown in FIG. 1, a three-dimensional image processing apparatus according to the present embodiment includes an original image memory 2 for storing a three-dimensional image obtained by a magnetic resonance tomography apparatus (not shown) and the like. An image processing unit 1 for performing image processing on the image signal in the direction, an input unit 3 for inputting an intensity conversion function used in the image processing unit 1, and an intensity conversion function for storing the intensity conversion function input by the input unit 3 A memory 4 and a display unit 5 for displaying a three-dimensional image processed by the image processing unit 1 are provided.

【0010】原画像メモリ2には、磁気共鳴断層撮影装置(以下、単に「MRI装置等」と呼ぶ)などで撮像された被撮像対象の複数の断層画像に基づく3次元画像が記憶されている。 [0010] The original image memory 2 stores a three-dimensional image based on a plurality of tomographic images of an object to be imaged, which are imaged by a magnetic resonance tomography apparatus (hereinafter simply referred to as "MRI apparatus"). .

【0011】具体的には、図2に示すように、MRI装置等によって、被撮像対象Mのn箇所の断層像を撮像することにより、例えば512×512画素の断層画像K
1〜Knがそれぞれ得られる。 1 to Kn are obtained respectively. 各断層画像K1〜Knには、512×512画素内に被撮像対象Mの断層像が含まれている。 Each tomographic image K1 to Kn includes a tomographic image of the image to be imaged M in 512 × 512 pixels. このようにして得られた断層画像K1〜K Tomographic images K1 to K obtained in this way
nは、全体として被撮像対象Mの3次元情報を含んでおり、各断層画像K1〜Knを積み重ねることで、例えば図3に示すように、512×512×n画素の3次元画像が得られる。 n includes the three-dimensional information of the image to be imaged M as a whole, and by stacking the tomographic images K1 to Kn, for example, as shown in FIG. 3, a three-dimensional image of 512 × 512 × n pixels can be obtained. .. 原画像メモリ2には、この512×51 The original image memory 2 has this 512 × 51
2×n画素の3次元画像が記憶される。 A 3D image of 2 × n pixels is stored. なお、本発明は、本実施例で説明する画素数の画像に限定されるものではない。 The present invention is not limited to the image having the number of pixels described in this embodiment. Specifically, as shown in FIG. 2, an MRI apparatus or the like captures n tomographic images of an object M to be imaged, for example, a tomographic image K of 512 × 512 pixels. Specifically, as shown in FIG. 2, an MRI apparatus or the like captures n tomographic images of an object M to be imaged, for example, a tomographic image K of 512 x 512 pixels.
1 to Kn are obtained respectively. Each of the tomographic images K1 to Kn includes a tomographic image of the imaging target M in 512 × 512 pixels. The tomographic images K1 to K obtained in this manner 1 to Kn are obtained respectively. Each of the tomographic images K1 to Kn includes a tomographic image of the imaging target M in 512 x 512 pixels. The tomographic images K1 to K obtained in this manner
n includes the three-dimensional information of the imaging target M as a whole, and by stacking the tomographic images K1 to Kn, for example, a three-dimensional image of 512 × 512 × n pixels is obtained as shown in FIG. . In the original image memory 2, the 512 × 51 n includes the three-dimensional information of the imaging target M as a whole, and by stacking the tomographic images K1 to Kn, for example, a three-dimensional image of 512 x 512 x n pixels is obtained as shown in FIG. the original image memory 2, the 512 x 51
A three-dimensional image of 2 × n pixels is stored. The present invention is not limited to the image having the number of pixels described in this embodiment. A three-dimensional image of 2 × n pixels is stored. The present invention is not limited to the image having the number of pixels described in this embodiment.

【0012】入力部3は、キーボードやマウスなどの入
力装置で構成されており、後述する強度変換関数を入力
するものである。入力部3から入力された強度変換関数
は、強度変換メモリ4に記憶される。
The input unit 3 is composed of an input device such as a keyboard and a mouse, and inputs an intensity conversion function described later. The intensity conversion function input from the input unit 3 is stored in the intensity conversion memory 4.

【0013】表示部5は、CRTモニタや液晶モニタな
どの表示装置で構成されており、画像処理部1で画像処
理された3次元画像や、3次元画像の断層画像などを表
示するものである。
The display unit 5 is composed of a display device such as a CRT monitor or a liquid crystal monitor, and displays a three-dimensional image processed by the image processing unit 1, a tomographic image of the three-dimensional image, and the like. .

【0014】画像処理部1は、3次元画像にウェーブレ
ット変換を施し、3次元画像の画像信号を周波数帯域ご
との成分である多重解像度成分に分解して、3次元画像
の低周波成分である近似成分と、高周波成分である複数
の詳細成分とを求める多重解像度分解部11と、多重解
像度分解部11で求められた近似成分と、複数の詳細成
分とを記憶する多重解像度成分メモリ12と、強度変換
関数に基づいて、詳細成分に強度変換を施す詳細成分強
度変換部13と、強度変換された詳細細分を含む複数の
詳細成分と、近似成分とで構成される強度変換後多重解
像度成分にウェーブレット逆変換を施して、3次元画像
を再構成する画像再構成部14と、再構成された3次元
画像を記憶する再構成画像メモリ15とを備えて構成さ
れている。なお、画像処理部1は、CPUやプログラム
メモリなどを備えるいわゆるコンピュータシステムで構
成されるものである。また、画像処理部11は本発明に
おける画像処理手段に、詳細成分強度変換部13は本発
明における強度変換手段に、画像再構成部14は本発明
における画像再構成手段に、それぞれ相当する。
The image processing unit 1 performs a wavelet transform on the three-dimensional image, decomposes the image signal of the three-dimensional image into multi-resolution components that are components for each frequency band, and approximates a low-frequency component of the three-dimensional image. A multi-resolution decomposition unit 11 for obtaining components and a plurality of detailed components which are high frequency components; a multi-resolution component memory 12 for storing the approximate components obtained by the multi-resolution decomposition unit 11 and a plurality of detailed components; A detailed component intensity conversion unit 13 for performing intensity conversion on the detailed component based on the conversion function; a plurality of detailed components including the detailed subdivisions subjected to the intensity conversion; It comprises an image reconstruction unit 14 for performing a reverse transformation to reconstruct a three-dimensional image, and a reconstructed image memory 15 for storing a reconstructed three-dimensional image. Note that the image processing unit 1 is configured by a so-called computer system including a CPU, a program memory, and the like. The image processing unit 11 corresponds to an image processing unit in the present invention, the detailed component intensity converting unit 13 corresponds to an intensity converting unit in the present invention, and the image reconstructing unit 14 corresponds to an image reconstructing unit in the present invention.

【0015】多重解像度分解部11は、3次元画像の画
像信号を高周波から低周波までの周波数帯域ごとの成分
に段階的に分解するウェーブレット変換を行なうもので
ある。このウェーブレット変換は、所望の基底関数に基
づいて、画像信号に畳み込み演算を行なうことによっ
て、画像信号を高周波成分である詳細成分と、低周波成
分である近似成分とに分解するものであり、例えば「ウ
ェーブレットビギナーズガイド」(榊原進著,1995
年5月20日,東京電機大学出版局発行)に開示されて
いる周知の手法である。以下、理解容易のため、Haa
rの基底関数を利用したウェーブレット変換の原理について、図4を参照しながら簡単に説明する。 The principle of the wavelet transform using the basis function of r will be briefly described with reference to FIG. なお、3次元画像の画像信号を周波数帯域ごとの成分である多重解像度成分に分解することを多重解像度分解といい、多重解像度成分の高周波成分を詳細成分といい、低周波成分を近似成分という。 Decomposing an image signal of a three-dimensional image into a multi-resolution component which is a component for each frequency band is called a multi-resolution decomposition, a high-frequency component of the multi-resolution component is called a detailed component, and a low-frequency component is called an approximate component. The multi-resolution decomposition section 11 performs a wavelet transform for gradually decomposing the image signal of the three-dimensional image into components for each frequency band from high frequency to low frequency. This wavelet transform is to decompose an image signal into a detailed component that is a high-frequency component and an approximate component that is a low-frequency component by performing a convolution operation on the image signal based on a desired basis function. "Wavelet Beginner's Guide" (Susumu Sakakibara, 1995 The multi-resolution decomposition section 11 performs a wavelet transform for gradually decomposing the image signal of the three-dimensional image into components for each frequency band from high frequency to low frequency. This wavelet transform is to decompose an image signal into a detailed component that "Wavelet Beginner's Guide" (Susumu Sakakibara, 1995) is a high-frequency component and an approximate component that is a low-frequency component by performing a convolution operation on the image signal based on a desired basis function.
May 20, 2010, published by Tokyo Denki University Press). Hereinafter, for easy understanding, Haa May 20, 2010, published by Tokyo Denki University Press). Increasing, for easy understanding, Haa
The principle of the wavelet transform using the basis function of r will be briefly described with reference to FIG. Decomposing an image signal of a three-dimensional image into multi-resolution components that are components for each frequency band is called multi-resolution decomposition, the high-frequency component of the multi-resolution component is called a detailed component, and the low-frequency component is called an approximate component. The principle of the wavelet transform using the basis function of r will be briefly described with reference to FIG. Decomposing an image signal of a three-dimensional image into multi-resolution components that are components for each frequency band is called multi-resolution decomposition, the high-frequency component of the multi-resolution component is called a detailed component, and the low-frequency component is called an approximate component.

【0016】図4(a)に示すように、例えば時間t=
0 〜t 7の間で、値「0」〜「6」で変化する画像信号C(t)に、Haarの基底関数の畳み込み演算を行なう。 The convolution operation of the basis function of Haar is performed on the image signal C (t) that changes with the values ​​"0" to "6" between t 0 and t 7 . これにより、画像信号C(t)は、図4(b)に示すように、画像信号C(t)の低周波成分である近似成分C (-1) (t)と、高周波成分である詳細成分D (- 1) As a result, as shown in FIG. 4B, the image signal C (t) is the approximate component C (-1) (t) which is a low frequency component of the image signal C (t) and the details which are high frequency components. Component D ( -1)
(t)とに分解される。 It is decomposed into (t). 以下、C (-1) (t)を単に「C Hereinafter, C (-1) (t) is simply referred to as "C."
-1 (t)」と、D (-1) (t)を単に「D -1 (t)」と表す。 -1 (t) ”and D (-1) (t) are simply expressed as“ D -1 (t) ”. As shown in FIG. 4A, for example, the time t = As shown in FIG. 4A, for example, the time t =
A convolution operation of a Haar basis function is performed on the image signal C (t) that changes between values “0” and “6” between t 0 and t 7 . Thus, as shown in FIG. 4B, the image signal C (t) has an approximate component C (-1) (t) which is a low frequency component of the image signal C (t) and details which is a high frequency component. Component D ( -1) A convolution operation of a Haar basis function is performed on the image signal C (t) that changes between values ​​“0” and “6” between t 0 and t 7. Thus, as shown in FIG. 4B, the image signal C ( t) has an approximate component C (-1) (t) which is a low frequency component of the image signal C (t) and details which is a high frequency component. Component D ( -1)
(T). Hereinafter, C (-1) (t) is simply referred to as “C (T). Incorporated, C (-1) (t) is simply referred to as “C
-1 (t) "and D (-1) (t) simply as" D -1 (t) ". -1 (t) "and D (-1) (t) simply as" D -1 (t) ".

【0017】近似成分C -1 (t)は、時間t 0,1の平均値である値「4」、時間t 2,3の平均値である値「4」、時間t 4,5の平均値である値「3」、時間t
6,7の平均値である値「1」によって表される信号成分である。 It is a signal component represented by a value "1" which is an average value of 6, t 7 . The approximate component C −1 (t) is a value “4” that is an average value of the times t 0 and t 1 , a value “4” that is an average value of the times t 2 and t 3 , and the times t 4 and t The value "3", which is the average value of 5 , the time t The approximate component C −1 (t) is a value “4” that is an average value of the times t 0 and t 1 , a value “4” that is an average value of the times t 2 and t 3 , and the times t 4 and t The value "3", which is the average value of 5 , the time t
6 is a signal component represented by the value "1" is an average value of t 7. 6 is a signal component represented by the value "1" is an average value of t 7.

【0018】詳細成分D -1 (t)は、近似成分C
-1 (t)の時間t 0,1の値「4」に対する、画像信号C(t)の時間t 0との差を示す値「1」、近似成分C

-1 (t)の時間t 2,3の値「4」に対する、画像信号C(t)の時間t 2との差を示す値「−2」、近似成分C -1 (t)の時間t 4,5の値「3」に対する、画像信号C(t)の時間t 4との差を示す値「0」、近似成分C -1 (t)の時間t 6,7の値「1」に対する、画像信号C(t)の時間t 6との差を示す値「0」によって表される信号成分である。 -1 (t) time t 2, t 3 value “4”, value “-2” indicating the difference between the image signal C (t) time t 2 and the approximate component C -1 (t) time The value "0" indicating the difference between the value "3" of t 4 and t 5 and the time t 4 of the image signal C (t), and the value "0" of the time t 6 and t 7 of the approximate component C -1 (t) It is a signal component represented by a value "0" indicating a difference between the image signal C (t) and the time t 6 with respect to "1". したがって、詳細成分D Therefore, the detailed component D
-1 (t)と、近似成分C -1 (t)とから、上述した逆の手順を行なえば、画像信号C(t)を導くことも可能である。 It is also possible to derive the image signal C (t) from -1 (t) and the approximate component C -1 (t) by performing the reverse procedure described above. The detailed component D -1 (t) is the approximate component C The detailed component D -1 (t) is the approximate component C
The value “1” indicating the difference between the time t 0 of the image signal C (t) and the value “4” of the time t 0 and t 1 of −1 (t), and the approximate component C The value “1” indicating the difference between the time t 0 of the image signal C (t) and the value “4” of the time t 0 and t 1 of −1 (t), and the approximate component C
-1 for (t) time t 2, t 3 values "4", a value indicating a difference between the time t 2 of the image signal C (t) "2", proximate component C -1 in (t) time to the value of t 4, t 5 "3", the value "0" indicating the difference between the time t 4 of the image signal C (t), the value of the approximation component C -1 time t 6, t 7 of (t) " This is a signal component represented by a value “0” indicating a difference from the time t 6 of the image signal C (t) with respect to “1”. Therefore, the detailed component D -1 for (t) time t 2, t 3 values ​​"4", a value indicating a difference between the time t 2 of the image signal C (t) "2", proximate component C -1 in (t) time to the value of t 4, t 5 "3", the value "0" indicating the difference between the time t 4 of the image signal C (t), the value of the approximation component C -1 time t 6, t 7 of (t) "This is a signal component represented by a value" 0 "indicating a difference from the time t 6 of the image signal C (t) with respect to" 1 ". Therefore, the detailed component D
By performing the above-described reverse procedure from -1 (t) and the approximate component C -1 (t), the image signal C (t) can be derived. By performing the above-described reverse procedure from -1 (t) and the approximate component C -1 (t), the image signal C (t) can be derived.

【0019】同様にして、低周波成分である近似成分C
-1 (t)に、基底関数の畳み込みを行なうことで、さらに、その低周波成分内における高周波成分と、低周波成分とに分解する。これにより、近似成分C -1 (t)は、
図4(c)に示すように、近似成分C -1 (t)の近似成分C -2 (t)と、詳細成分D -2 (t)とに分解される。
同様に、基底関数の畳み込みを繰り返し行なうことで、
近似成分C -n (t)と、複数の詳細成分D -1 (t)〜D

-n (t)とが求められる。 -n (t) is required. なお、近似成分に基底関数の畳み込みを繰り返すことで、画像信号に含まれる高周波帯域の信号成分から順番に分解される。 By repeating the convolution of the basis function into the approximate component, the signal component in the high frequency band included in the image signal is decomposed in order. その結果、多重解像度分解成分である、単一の近似成分と、複数の詳細成分とが求められる。 As a result, a single approximate component, which is a plural resolution decomposition component, and a plurality of detailed components are required. なお、ウェーブレット変換の特性から、近似成分と、詳細成分とである多重解像度成分に、上述した逆の手順を実施することで、さらに上位の近似成分を構成することができる。 From the characteristics of the wavelet transform, it is possible to construct a higher approximate component by performing the reverse procedure described above for the multiple resolution component which is the approximate component and the detailed component. つまり、元の画像を再現することができる。 That is, the original image can be reproduced. 以上がHaarの基底関数を利用したウェーブレット変換についての簡単な説明であるが、この説明はウェーブレット変換の原理説明であるので、一般の基底関数を利用した場合のウェーブレット変換の詳細については上述した文献「ウェーブレットビギナーズガイド」に記載されている。 The above is a brief explanation of the wavelet transform using the Haar basis function, but since this explanation is the explanation of the principle of the wavelet transform, the details of the wavelet transform when using a general basis function are described in the above-mentioned documents. It is described in "Wavelet Beginner's Guide". Similarly, the approximate component C which is a low frequency component Similarly, the approximate component C which is a low frequency component
By performing convolution of the basis function on -1 (t), it is further decomposed into a high-frequency component and a low-frequency component in the low-frequency component. Thus, the approximate component C -1 (t) is By performing convolution of the basis function on -1 (t), it is further decomposed into a high-frequency component and a low-frequency component in the low-frequency component. Thus, the approximate component C -1 (t) is
As shown in FIG. 4C, the approximate component C -1 (t) is decomposed into an approximate component C -2 (t) and a detailed component D -2 (t). As shown in FIG. 4C, the approximate component C -1 (t) is decomposed into an approximate component C -2 (t) and a detailed component D -2 (t).
Similarly, by repeatedly convolving the basis functions, Similarly, by repeatedly convolving the basis functions,
The approximate component C -n (t) and a plurality of detailed components D -1 (t) to D The approximate component C -n (t) and a plurality of detailed components D -1 (t) to D
-n (t). Note that by repeating convolution of the basis function with the approximate component, the image component is sequentially decomposed from the high-frequency band signal component included in the image signal. As a result, a single approximate component, which is a multi-resolution decomposition component, and a plurality of detailed components are obtained. In addition, from the characteristics of the wavelet transform, a higher-order approximate component can be configured by performing the above-described reverse procedure on the multi-resolution component that is the approximate component and the detailed component. That is, the original image can be reproduced. The above is a brief description of the wavelet transform using the Haar basis function. Since this description is a principle explanation of the wavelet transform, the details of the wavelet transform using the general basis function are described in the above-mentioned reference. It is described in the "Wavelet -n (t). Note that by repeating convolution of the basis function with the approximate component, the image component is sequentially decomposed from the high-frequency band signal component included in the image signal. As a result, a single approximate component, which In addition, from the characteristics of the wavelet transform, a higher-order approximate component can be configured by performing the above-described reverse procedure on the multi-resolution. is a multi-resolution decomposition component, and a plurality of detailed components are obtained. That is, the original image can be reproduced. The above is a brief description of the wavelet transform using the Haar basis function. Since this description is a principle explanation of the wavelet transform, the details of the wavelet transform using the general basis function are described in the above-mentioned reference. It is described in the "Wavelet Beginner's Guide". Beginner's Guide ".

【0020】多重解像度分解部11は、図5(a)に示
すように、原画像メモリ2に記憶されている3次元画像
のX方向(矢印で示す)の画像信号についてウェーブレ
ット変換を行なう。これにより、3次元画像の各断層画
像K1〜Knは、X方向が低周波成分である近似成分
と、X方向が高周波成分である詳細成分とに分解され
る。その結果、図5(b)に示すように、3次元画像
は、X方向が低周波成分である近似成分Lと、X方向が
高周波成分である詳細成分Hとに分解される。
As shown in FIG. 5A, the multi-resolution decomposition section 11 performs a wavelet transform on the image signal of the three-dimensional image stored in the original image memory 2 in the X direction (indicated by an arrow). Thereby, each tomographic image K1 to Kn of the three-dimensional image is decomposed into an approximate component in which the X direction is a low frequency component and a detailed component in which the X direction is a high frequency component. As a result, as shown in FIG. 5B, the three-dimensional image is decomposed into an approximate component L having a low frequency component in the X direction and a detailed component H having a high frequency component in the X direction.

【0021】また、多重解像度分解部11は、X方向に
ついて低周波成分と、高周波成分とに分解した3次元画
像のY方向の画像信号についてウェーブレット変換を行
なう(図5(b)の矢印参照)。これにより、3次元画
像の各断層画像K1〜Knは、X、Y方向の両方が低周
波成分である近似成分と、X、Y方向の両方が高周波成
分である詳細成分と、X方向が低周波成分でY方向が高
周波成分である詳細成分と、X方向が高周波成分でY方
向が低周波成分である詳細成分とに分解される。その結
果、図5(c)に示すように、3次元画像は、X、Y方
向の両方が低周波成分である近似成分LLと、X、Y方
向の両方が高周波成分である詳細成分HHと、X方向が
低周波成分でありY方向が高周波成分である詳細成分L
Hと、X方向が高周波成分でありY方向が低周波成分である詳細成分HLに分解される。 It is decomposed into H and a detailed component HL in which the X direction is a high frequency component and the Y direction is a low frequency component. The multiresolution decomposition section 11 performs a wavelet transform on the image signal in the Y direction of the three-dimensional image decomposed into the low frequency component and the high frequency component in the X direction (see the arrow in FIG. 5B). . Accordingly, each of the tomographic images K1 to Kn of the three-dimensional image has an approximate component in which both the X and Y directions are low frequency components, a detailed component in which both the X and Y directions are high frequency components, and a low component in the X direction. The frequency component is decomposed into a detailed component whose Y direction is a high frequency component and a detailed component whose X direction is a high frequency component and the Y direction is a low frequency component. As a result, as shown in FIG. 5C, the three-dimensional image has an approximate component LL in which both the X and Y directions are low frequency components and a detailed component H The multiresolution decomposition section 11 performs a wavelet transform on the image signal in the Y direction of the three-dimensional image decomposed into the low frequency component and the high frequency component in the X direction (see the arrow in FIG. 5B). , each of the tomographic images K1 to Kn of the three-dimensional image has an approximate component in which both the X and Y directions are low frequency components, a detailed component in which both the X and Y directions are high frequency components, and a low component in the X direction. The frequency component is decomposed into a detailed component whose Y direction is a high frequency component and a detailed component whose X direction is a high frequency component and the Y direction is a low frequency component. As a result, as shown in FIG. 5C, the three-dimensional image has an approximate component LL in which both the X and Y directions are low frequency components and a detailed component H H in which both the X and Y directions are high frequency components. , A detailed component L in which the X direction is a low frequency component and the Y direction is a high frequency component H in which both the X and Y directions are high frequency components., A detailed component L in which the X direction is a low frequency component and the Y direction is a high frequency component
It is decomposed into H and a detailed component HL in which the X direction is a high frequency component and the Y direction is a low frequency component. It is decomposed into H and a detailed component HL in which the X direction is a high frequency component and the Y direction is a low frequency component.

【0022】さらに、多重解像度分解部11は、X、Y
方向について低周波成分と、高周波成分とに分解した3
次元画像のZ方向の画像信号についてウェーブレット変換を行なう(図5(c)の矢印参照)。これにより、3

次元画像のZ方向の画像信号が、低周波成分と、高周波成分とに分解される。 The image signal in the Z direction of the dimensional image is decomposed into a low frequency component and a high frequency component. その結果、図5(d)に示すように、3次元画像は、X、Y、Z方向が低周波成分である近似成分LLLと、X、Y、Z方向が高周波成分である詳細成分HHHと、X、Y、Z方向のいずれか1つが低周波成分である詳細成分LHH、HLH、HHLと、 As a result, as shown in FIG. 5D, in the three-dimensional image, the approximate component LLL whose X, Y, and Z directions are low frequency components and the detailed component HHH whose X, Y, and Z directions are high frequency components are , X, Y, Z direction is a low frequency component, detailed components LHH, HLH, HHL, and
X、Y、Z方向のいずれか2つが低周波成分である詳細成分LLH、HLL、LHLとに分解される。 Any two of the X, Y, and Z directions are decomposed into detailed components LLH, HLL, and LHL, which are low-frequency components. なお、上述したウェーブレット変換の簡単な説明に対応させた符号を用いると、3次元画像の各方向に1回のみウェーブレット変換を施しているので、近似成分LLLはLLL If a code corresponding to the above-mentioned brief description of the wavelet transform is used, the wavelet transform is performed only once in each direction of the three-dimensional image, so that the approximate component LLL is LLL.
(-1)と表すことができ、各詳細成分HHH、LHH、H It can be expressed as (-1), and each detailed component HHH, LHH, H
LH、HHL、LLH、HLL、LHLは、各詳細成分HHH (-1) 、LHH (-1) 、HLH (-1) 、HHL (-1) 、L LH, HHL, LLH, HLL, LHL are the detailed components HHH (-1) , LHH (-1) , HLH (-1) , HHL (-1) , L.
LH (-1) 、HLL (-1) 、LHL (-1)と表すことができる。 It can be expressed as LH (-1) , HLL (-1) , LHL (-1) . なお、HHH (-1) 、・・・・、LLL (-1)を単に、 In addition, HHH (-1) , ..., LLL (-1) is simply referred to as
「HHH -1 、・・・・、LLL -1 」と示すが、以下の「 Although it is indicated as "HHH -1 , ..., LLL -1 ", the following "
-1 」、「 -2 」・・・「 -n 」はいわゆる指数ではなく、ウェーブレット変換を行なった回数を示すものである。 -1 "," -2 "..." -n "is not a so-called exponent, but indicates the number of wavelet transforms performed. Further, the multi-resolution decomposition section 11 performs X, Y Further, the multi-resolution decomposition section 11 performs X, Y
Decomposed into low frequency component and high frequency component in the direction 3 Decomposed into low frequency component and high frequency component in the direction 3
Wavelet transform is performed on the image signal of the two-dimensional image in the Z direction (see the arrow in FIG. 5C). This gives 3 Wavelet transform is performed on the image signal of the two-dimensional image in the Z direction (see the arrow in FIG. 5C). This gives 3
An image signal in the Z direction of the two-dimensional image is decomposed into a low-frequency component and a high-frequency component. As a result, as shown in FIG. 5D, the three-dimensional image has an approximate component LLL in which the X, Y, and Z directions are low-frequency components, and a detailed component HHH in which the X, Y, and Z directions are high-frequency components. , XH, HLH, and HHL, which are low-frequency components in one of the X, Y, and Z directions; An image signal in the Z direction of the two-dimensional image is decomposed into a low-frequency component and a high-frequency component. As a result, as shown in FIG. 5D, the three-dimensional image has an approximate component LLL in which the X, Y, and Z directions are low-frequency components, and a detailed component HHH in which the X, Y, and Z directions are high-frequency components., XH, HLH, and HHL, which are low-frequency components in one of the X, Y, and Z directions;
Any two of the X, Y, and Z directions are decomposed into detailed components LLH, HLL, and LHL, which are low frequency components. When a code corresponding to the above brief description of the wavelet transform is used, since the wavelet transform is performed only once in each direction of the three-dimensional image, the approximate component LLL is LLL. Any two of the X, Y, and Z directions are decomposed into detailed components LLH, HLL, and LHL, which are low frequency components. When a code corresponding to the above brief description of the wavelet transform is used, since the wavelet transform is performed only once in each direction of the three-dimensional image, the approximate component LLL is LLL.
(-1), and each detailed component HHH, LHH, H (-1), and each detailed component HHH, LHH, H
LH, HHL, LLH, HLL, LHL are the detailed components HHH (-1) , LHH (-1) , HLH (-1) , HHL (-1) , L LH, HHL, LLH, HLL, LHL are the detailed components HHH (-1) , LHH (-1) , HLH (-1) , HHL (-1) , L
LH (-1) , HLL (-1) and LHL (-1) . Note that HHH (-1) ,..., LLL (-1) is simply expressed as LH (-1) , HLL (-1) and LHL (-1) . Note that HHH (-1) , ..., LLL (-1) is simply expressed as
Although “HHH −1 ,..., LLL −1 ” is shown, the following “ Although “HHH −1 , ..., LLL −1 ” is shown, the following “
−1 ”, “ −2,. "-1", "-2",.

【0023】多重解像度分解部11は、同様にして、近似成分LLL -1のX、Y、Zの各方向にウェーブレット変換を施すことで、近似成分LLL -1をさらに、高周波成分と低周波成分に分解する。これにより、3次元画像の近似成分LLL -1は、近似成分LLL -2 、詳細成分H
HH -2 、LHH -2 、HLH -2 、HHL -2 、LLH -2 、H HH -2 , LHH -2 , HLH -2 , HHL -2 , LLH -2 , H
LL -2 、LHL -2に分解される(図5(e)の概略図参照)。 It is decomposed into LL- 2 and LHL- 2 (see the schematic diagram of FIG. 5 (e)). その結果、3次元画像は、近似成分LLL -2 、詳細成分HHH -2 、LHH -2 、HLH -2 、HHL As a result, the three-dimensional image has an approximate component LLL- 2 , a detailed component HHH- 2 , LHH- 2 , HLH- 2 , and HHL. -2 、LL -2 , LL
-2 、HLL -2 、LHL -2と、詳細成分HHH -1 、LH H- 2 , HLL- 2 , LHL- 2 and detailed components HHH- 1 , LH
-1 、HLH -1 、HHL -1 、LLH -1 、HLL -1 、LH H -1 , HLH -1 , HHL -1 , LLH -1 , HLL -1 , LH
-1とで構成される多重解像度成分が得られる。 A multi-resolution component composed of L -1 is obtained. さらに、近似成分についてのウェーブレット変換を繰り返すことにより、より低い低周波帯域と高周波帯域との信号成分に分解することができる。 Furthermore, by repeating the wavelet transform on the approximate component, it can be decomposed into the signal components of the lower low frequency band and the high frequency band. つまり、近似成分に対してウェーブレット変換をn回施すことにより、近似成分LLL -n 、詳細成分HHH -n 、・・・・を得ることができる。 That is, the approximate component LLL -n , the detailed component HHH -n , ... Can be obtained by performing the wavelet transform on the approximate component n times. The multi-resolution decomposition section 11 similarly operates The multi-resolution decomposition section 11 similarly operates
Similar component LLL -1 Wavelets in each of X, Y and Z directions Similar component LLL -1 Wavelets in each of X, Y and Z directions
By performing the conversion, the approximate component LLL -1 The further high frequency By performing the conversion, the approximate component LLL -1 The further high frequency
Decomposes into components and low frequency components. This gives a 3D image Decomposes into components and low frequency components. This gives a 3D image
Approximate component LLL -1 Is the approximate component LLL -2 , Detailed ingredient H Approximate component LLL -1 Is the approximate component LLL -2 , Detailed ingredient H
HH -2 , LHH -2 , HLH -2 , HHL -2 , LLH -2 , H HH -2 , LHH -2 , HLH -2 , HHL -2 , LLH -2 , H
LL -2 , LHL -2 (See the schematic diagram in FIG. 5 (e)). LL -2 , LHL -2 (See the schematic diagram in FIG. 5 (e)).
See). As a result, the three-dimensional image is represented by the approximate component LLL. -2 , Details See). As a result, the three-dimensional image is represented by the approximate component LLL. -2 , Details
Fine component HHH -2 , LHH -2 , HLH -2 , HHL Fine component HHH -2 , LHH -2 , HLH -2 , HHL -2 , LL -2 , LL
H -2 , HLL -2 , LHL -2 And the detailed component HHH -1 , LH H -2 , HLL -2 , LHL -2 And the detailed component HHH -1 , LH
H -1 , HLH -1 , HHL -1 , LLH -1 , HLL -1 , LH H -1 , HLH -1 , HHL -1 , LLH -1 , HLL -1 , LH
L -1 Is obtained. Further L -1 Is obtained. Further
And repeat the wavelet transform for the approximate component And repeat the wavelet transform for the approximate component
Signal in the lower and higher frequency bands Signal in the lower and higher frequency bands
Can be broken down into components. In other words, for the approximate component Can be broken down into components. In other words, for the approximate component
By applying the wavelet transform n times By applying the wavelet transform n times
LLL -n , Detailed components HHH -n ... LLL -n , Detailed components HHH -n ...
Wear. Wear.

【0024】多重解像度成分メモリ12は、多重解像度
分解部11で求められた、例えば図5(e)の概略図で
示す3次元画像の多重解像度成分である、近似成分LL
-2と、詳細成分HHH-2、LHH-2、HLH-2、HH
-2、LLH-2、HLL-2、LHL-2と、詳細成分HH
-1、LHH-1、HLH-1、HHL-1、LLH-1、HL
-1、LHL-1とを記憶する。
The multi-resolution component memory 12 stores an approximate component LL, which is a multi-resolution component of the three-dimensional image shown in the schematic diagram of FIG.
L- 2 and detailed components HHH- 2 , LHH- 2 , HLH- 2 , HH L- 2 and detailed components HHH- 2 , LHH- 2 , HLH- 2 , HH
L- 2 , LLH- 2 , HLL- 2 , LHL- 2 , and detailed component HH L- 2 , LLH- 2 , HLL- 2 , LHL- 2 , and detailed component HH
H −1 , LHH −1 , HLH −1 , HHL −1 , LLH −1 , HL H -1 , LHH -1 , HLH -1 , HHL -1 , LLH -1 , HL
L -1 and LHL -1 are stored. L -1 and LHL -1 are stored.

【0025】詳細成分強度変換部13は、多重解像度成分メモリ12から例えば詳細成分HHH -2 、LHH -2
HLH -2 、HHL -2 、LLH -2 、HLL -2 、LHL -2を取り出して、これら各詳細成分に強度変換を施す。 HLH- 2 , HHL- 2 , LLH- 2 , HLL- 2 , and LHL- 2 are taken out, and each of these detailed components is subjected to intensity conversion. この強度変換は、入力部3で設定された強度変換関数に基づいて行なわれる。 This intensity conversion is performed based on the intensity conversion function set in the input unit 3. なお、本実施例では、上述した詳細成分に強度変換を施すが、本発明はこれに限定されるものではなく、任意の詳細成分について強度変換を施したり、各段階の詳細成分に各々異なる強度変換を施すこともできる。 In this embodiment, the above-mentioned detailed components are subjected to strength conversion, but the present invention is not limited to this, and any detailed component may be subjected to strength conversion, or the detailed components at each stage may have different strengths. It can also be converted. The detailed component intensity converter 13 outputs, for example, the detailed components HHH -2 , LHH -2 , The detailed component intensity converter 13 outputs, for example, the detailed components HHH -2 , LHH -2 ,
HLH -2 , HHL -2 , LLH -2 , HLL -2 , and LHL -2 are extracted and subjected to intensity conversion for each of these detailed components. This intensity conversion is performed based on the intensity conversion function set by the input unit 3. In the present embodiment, the above-described detailed components are subjected to intensity conversion. However, the present invention is not limited to this. For example, the detailed components may be subjected to intensity conversion, or the detailed components at each stage may have different intensities. Conversion can also be performed. HLH -2 , HHL -2 , LLH -2 , HLL -2 , and LHL -2 are extracted and subjected to intensity conversion for each of these detailed components. This intensity conversion is performed based on the intensity conversion function set by the input unit 3. In the present embodiment, the above-described detailed components are subjected to intensity conversion. However, the present invention is not limited to this. For example, the detailed components may be subjected to intensity conversion, or the detailed components at each stage May have different intensities. Conversion can also be performed.

【0026】以下、ウェーブレット変換を2回施した2
段階の詳細成分の場合について以下説明する。 The case of the detailed component of the step will be described below. なお、本発明は、2回のウェーブレット変換に限定されるものではなく、1回以上のウェーブレット変換に適用することができる。 The present invention is not limited to two wavelet transforms, and can be applied to one or more wavelet transforms. Hereinafter, the wavelet transform is performed twice. TWICE, the wavelet transform is performed twice.
The case of the detailed component of the stage will be described below. Note that the present invention is not limited to two wavelet transforms, but can be applied to one or more wavelet transforms. The case of the detailed component of the stage will be described below. Note that the present invention is not limited to two wavelet transforms, but can be applied to one or more wavelet transforms.

【0027】この場合、例えば、強度変換関数は次のように設定される。まず、図6に示すように、3次元画像内に被撮像対象M1と、雑音成分S1、S2とが含まれており、被撮像対象M1の断面を観察したい場合には、
被撮像対象M1の断面成分を強調させるとともに、雑音成分S1、S2を低減させる必要がある。 It is necessary to emphasize the cross-sectional component of the image target M1 and reduce the noise components S1 and S2. この場合に、 In this case,
3次元画像の例えば断層画像KiのプロファイルラインPには、図7に示すように、被撮像対象M1の断層成分では比較的高濃度値をもち、雑音成分S1、S2では比較的低濃度値をもつ画像信号Ciが現れる。 As shown in FIG. 7, the profile line P of, for example, the tomographic image Ki of the three-dimensional image has a relatively high density value in the tomographic component of the imaged object M1 and a relatively low density value in the noise components S1 and S2. Image signal Ci appears. In this case, for example, the intensity conversion function is set as follows. First, as shown in FIG. 6, when a three-dimensional image includes an imaging target M1 and noise components S1 and S2, and it is desired to observe a cross section of the imaging target M1, In this case, for example, the intensity conversion function is set as follows. First, as shown in FIG. 6, when a three-dimensional image includes an imaging target M1 and noise components S1 and S2, and it is desired to observe a cross section of the imaging target M1,
It is necessary to emphasize the cross-sectional components of the imaging target M1 and reduce the noise components S1 and S2. In this case, It is necessary to emphasize the cross-sectional components of the imaging target M1 and reduce the noise components S1 and S2. In this case,
As shown in FIG. 7, a profile line P of a three-dimensional image, for example, a tomographic image Ki has a relatively high density value for the tomographic component of the imaging target M1, and a relatively low density value for the noise components S1 and S2. Image signal Ci appears. As shown in FIG. 7, a profile line P of a three-dimensional image, for example, a tomographic image Ki has a relatively high density value for the tomographic component of the imaging target M1, and a relatively low density value for the noise components S1 and S2. Image signal Ci appears.

【0028】画像信号Ciにウェーブレット変換を施した場合には、図8に示すように、雑音成分S1、S2での濃度変化と、被撮像対象M1の断面部での濃度変化が顕著に現れた詳細成分Di -l (l=1、2)が求まる。
以下、顕著に現れた詳細成分Di -1として説明する。 Hereinafter, the detailed component Di -1 that appears prominently will be described. ここで、雑音成分S1、S2の詳細成分Di -1での濃度変化の値をD1として、被撮像対象M1の断面部での値をD2とする。 Here, let D1 be the value of the density change in the detailed component Di -1 of the noise components S1 and S2, and let D2 be the value in the cross section of the object to be imaged M1. When the wavelet transform is applied to the image signal Ci, as shown in FIG. 8, the density change in the noise components S1 and S2 and the density change in the cross section of the object M1 appear remarkably. The detailed component Di -l (l = 1, 2) is obtained. When the wavelet transform is applied to the image signal Ci, as shown in FIG. 8, the density change in the noise components S1 and S2 and the density change in the cross section of the object M1 appear remarkably. The detailed component Di -l (l = 1, 2) is obtained.
Hereinafter, description will be given as a prominent detail component Di- 1 . Here, it is assumed that the value of the density change of the noise components S1 and S2 in the detailed component Di -1 is D1, and the value of the cross section of the imaging target M1 is D2. Similarly, description will be given as a prominent detail component Di- 1 . Here, it is assumed that the value of the density change of the noise components S1 and S2 in the detailed component Di -1 is D1, and the value of the cross section of the imaging target M1 is D2.

【0029】ここでは、雑音成分S1、S2の値D1を
小さくするとともに、被撮像対象M1の値D2を大きく
するような関数を設定すればよい。そこで、図9に示す
ように、値D1までは、その大きさが最も小さくなるよ
うな倍率a1が選ばれ、値D2以上では、その大きさが
最も大きくなるような倍率a2が選ばれ、値D1から値
D2の間では、倍率a1と倍率a2との中間の倍率が選
ばれるような強度変換関数Aを設定する。この強度変換
関数Aは、強度変換関数メモリ4に記憶される。
Here, a function may be set to reduce the value D1 of the noise components S1 and S2 and increase the value D2 of the object M1. Therefore, as shown in FIG. 9, a magnification a1 whose size is the smallest is selected up to the value D1, and a magnification a2 whose size is the largest is selected above the value D2. Between D1 and D2, an intensity conversion function A is set such that an intermediate magnification between magnifications a1 and a2 is selected. This intensity conversion function A is stored in the intensity conversion function memory 4.

【0030】なお、詳細成分Di-1に、強度変換関数A
を作用させることで、詳細成分Di -1の値に応じた倍率
がその値に乗じられ(例えば、a1×D1)、詳細成分
Di -1の強度が変換される。その結果、図10に示すよ
うに、雑音成分S1、S2の値D1が値「a1×D1」
にまで縮小され、被撮像対象M1の値D2が値「a2×
D2」にまで強調された強度変換後詳細成分Di-1’を
求めることができる。同様に、詳細成分Di-2について
も強度変換を行なう。
The detailed component Di-1The intensity conversion function A
To make the detailed component Di -1Magnification according to the value of
Is multiplied by the value (for example, a1 × D1), and the detailed component Is multiplied by the value (for example, a1 x D1), and the detailed component
Di Di -1 Is converted. As a result, as shown in FIG. -1 Is converted. As a result, as shown in FIG.
Thus, the value D1 of the noise components S1 and S2 is the value "a1 × D1". Thus, the value D1 of the noise components S1 and S2 is the value "a1 x D1".
And the value D2 of the object to be imaged M1 becomes the value “a2 × And the value D2 of the object to be imaged M1 becomes the value “a2 ×
D2 ”, the intensity-converted detailed component Di emphasized to“ D2 ” -1 ' D2 ”, the intensity-converted detailed component Di emphasized to“ D2 ” -1 '
You can ask. Similarly, the detailed component Di -2 about You can ask. Similarly, the detailed component Di -2 about
Also perform intensity conversion. Also perform intensity conversion.

【0031】詳細成分強度変換部13は、詳細成分HH
-2 、LHH -2 、HLH -2 、HHL -2 、LLH -2 、HL

-2 、LHL -2に対して、強度変換関数Aを作用させることで、雑音成分を低減させ、かつ、被撮像対象Mの例えば断面部が強調させた、強度変換後詳細成分HH By applying the intensity conversion function A to L- 2 and LHL- 2 , the noise component is reduced, and the detailed component HH after intensity conversion, for example, the cross section of the imaged object M is emphasized.
-2 '、LHH -2 '、HLH -2 '、HHL -2 '、LLH H -2 ', LHH -2 ', HLH -2 ', HHL -2 ', LLH
-2 '、HLL -2 '、LHL -2 'を求める。 Find -2 ', HLL -2 ', LHL -2 '. なお、各段階の詳細成分HHH -1等についても同様に、詳細成分HH Similarly, for the detailed component HHH -1 and the like at each stage, the detailed component HH
-1 '等を求める。 Find H -1'etc . The detailed component intensity converter 13 calculates the detailed component HH The detailed component intensity converter 13 calculates the detailed component HH
H -2 , LHH -2 , HLH -2 , HHL H -2 , LHH -2 , HLH -2 , HHL -2 , LLH -2 , HL -2 , LLH -2 , HL
L -2 , LHL -2 To the intensity conversion function A L -2 , LHL -2 To the intensity conversion function A
In this way, the noise component is reduced, and In this way, the noise component is reduced, and
For example, the detailed component HH after the intensity conversion in which the cross section is emphasized For example, the detailed component HH after the intensity conversion in which the cross section is emphasized
H -2 ', LHH -2 ', HLH -2 ', HHL -2 ', LLH H -2 ', LHH -2 ', HLH -2 ', HHL -2 ', LLH
-2 ', HLL -2 ', LHL -2 '. Each stage -2 ', HLL -2 ', LHL -2 '. Each stage
Detailed ingredient of HHH -1 Similarly, the detailed component HH Detailed ingredient of HHH -1 Similarly, the detailed ingredient HH
H -1 'And so on. H -1 'And so on.

【0032】画像再構成部14は、まず、近似成分LL
-2と、強度変換後詳細成分HHH -2 '、LHH -2 '、
HLH -2 '、HHL -2 '、LLH -2 '、HLL -2 '、L
HL -2 'とで構成される強度変換後多重解像度成分にウェーブレット逆変換を施して、強度変換後の近似成分L

LL -1 'を再構成する。 Reconstruct LL -1 '. さらに、画像再構成部14は、 Further, the image reconstruction unit 14
強度変化後の近似成分LLL -1 'と、詳細成分HH Approximate component LLL -1 'after intensity variation, detail component HH
-1 、LHH -1 、HLH -1 、HHL -1 、LLH -1 、HL H -1 , LHH -1 , HLH -1 , HHL -1 , LLH -1 , HL
-1 、LHL -1とにウェーブレット逆変換を施すことで、強度変換された3次元画像を再構成する。 By applying the wavelet inverse transform to L -1 and LHL -1 , the intensity-transformed three-dimensional image is reconstructed. この再構成された3次元画像は、雑音成分が低減されるとともに、所定成分である例えば断面成分が強調されたものである。 In this reconstructed three-dimensional image, the noise component is reduced and the predetermined component, for example, the cross-sectional component is emphasized. First, the image reconstruction unit 14 calculates the approximate component LL First, the image reconstruction unit 14 calculates the approximate component LL
L -2 And the detailed component HHH after the intensity conversion L -2 And the detailed component HHH after the intensity conversion -2 ', LHH -2 ', -2 ', LHH -2 ',
HLH -2 ', HHL -2 ', LLH -2 ', HLL -2 ', L HLH -2 ', HHL -2 ', LLH -2 ', HLL -2 ', L
HL HL -2 ′ To the multi-resolution component -2 ′ To the multi-resolution component
Approximate component L after intensity conversion by performing inverse wavelet transformation Approximate component L after intensity conversion by performing inverse wavelet transformation
LL -1 '. Further, the image reconstruction unit 14 LL -1 '. Further, the image reconstruction unit 14
Approximate component LLL after intensity change -1 'And the detailed component HH Approximate component LLL after intensity change -1 ' And the detailed component HH
H -1 , LHH -1 , HLH -1 , HHL -1 , LLH -1 , HL H -1 , LHH -1 , HLH -1 , HHL -1 , LLH -1 , HL
L -1 , LHL -1 And the inverse wavelet transform L -1 , LHL -1 And the inverse wavelet transform
Reconstructs the intensity-converted three-dimensional image. This reconstruction Reconstructs the intensity-converted three-dimensional image. This reconstruction
The resulting three-dimensional image has reduced noise components The resulting three-dimensional image has reduced noise components
In addition, a predetermined component, for example, a cross-sectional component is emphasized. In addition, a predetermined component, for example, a cross-sectional component is emphasized.
is there. is there.

【0033】画像再構成部14で再構成された3次元画像は、再構成画像メモリ15に記憶される。表示部5
は、再構成画像メモリ15に記憶された3次元画像を表示する。 Displays a three-dimensional image stored in the reconstructed image memory 15. この3次元画像を観察するオペレータは、従来と同様に、適宜任意断面を取り出して、3次元画像内の断面を把握する。 The operator observing the three-dimensional image appropriately takes out an arbitrary cross section and grasps the cross section in the three-dimensional image as in the conventional case. The three-dimensional image reconstructed by the image reconstruction unit 14 is stored in a reconstructed image memory 15. Display 5 The three-dimensional image reconstructed by the image reconstruction unit 14 is stored in a reconstructed image memory 15. Display 5
Displays the three-dimensional image stored in the reconstructed image memory 15. The operator who observes the three-dimensional image takes out an arbitrary cross section as appropriate and grasps the cross section in the three-dimensional image as in the related art. Displays the three-dimensional image stored in the reconstructed image memory 15. The operator who observes the three-dimensional image takes out an arbitrary cross section as appropriate and grasps the cross section in the three-dimensional image as in the related art.

【0034】上述した実施例装置では、ウェーブレット
変換によって画像処理を施しているので、一般的な空間
フィルタやフーリェ変換を利用した方法に比べて、空間
分解能と低周波の情報取得とが両立する限界が高く、高
周波から低周波帯域までの濃度解析を高精度に行なうこ
とができる。また、3次元データ構造を有する3次元画
像に、3次元空間内の3方向からウェーブレット変換を
施しているので、3次元画像に含まれる3次元情報を全
て活用することができる。その結果、被撮像対象を撮像
して得られる2次元画像間にまたがる3次元の雑音成分
を低減させたり、所定成分を強調させたりすることがで
きる。
In the above-described embodiment, the image processing is performed by the wavelet transform, so that the spatial resolution and the acquisition of low-frequency information are both compatible as compared with a general method using a spatial filter or a Fourier transform. And the density analysis from the high frequency to the low frequency band can be performed with high accuracy. In addition, since the three-dimensional image having the three-dimensional data structure is subjected to the wavelet transform from three directions in the three-dimensional space, all the three-dimensional information included in the three-dimensional image can be utilized. As a result, it is possible to reduce a three-dimensional noise component extending between two-dimensional images obtained by imaging the imaging target, or enhance a predetermined component.

【0035】この発明は上述した実施例に限られるもの
ではなく、以下のように変形実施することが可能であ
る。 (1)上述した実施例では、水平方向の複数の断層画像
によって構成された3次元画像について説明したが、本
発明はこれに限定されることなく、3次元のデータ構造
を備えた3次元画像であれば適用することができる。
The present invention is not limited to the above-described embodiment, but can be modified as follows. (1) In the above-described embodiment, a three-dimensional image constituted by a plurality of horizontal tomographic images has been described. However, the present invention is not limited to this, and a three-dimensional image having a three-dimensional data structure is provided. If so, it can be applied.

【0036】(2)上述した実施例では、2次元画像で
ある複数の断層画像によって構成された3次元画像の各
画像信号について画像処理を施したが、例えば3次元画
像をVoxel構造の3次元データとしても同様に画像
処理を施すことができる。
(2) In the above-described embodiment, image processing is performed on each image signal of a three-dimensional image composed of a plurality of tomographic images, which is a two-dimensional image. Image processing can be similarly performed on data.

【0037】(3)上述した実施例では、図9に示した
ような強度変換関数Aについて説明したが、この強度変
換関数Aは、適宜その関数の形状を設定することができ
るものであり、例えば、図11に示すように、値D3以
上では、その値の倍率を変化させないような強度変換関
数Aを設定することもできる。また、ウェーブレット変
換の回数によって、異なる形状の強度変換関数を設定す
るようにしてもよい。
(3) In the above-described embodiment, the intensity conversion function A as shown in FIG. 9 has been described. However, the intensity conversion function A can appropriately set the shape of the function. For example, as shown in FIG. 11, when the value is equal to or more than D3, an intensity conversion function A that does not change the magnification of the value can be set. Further, an intensity conversion function having a different shape may be set according to the number of wavelet transforms.

【0038】[0038]

【発明の効果】以上の説明から明らかなように、この発
明によれば、3次元画像の各次元方向の全ての情報を利
用して画像処理を施しているので、3次元画像全体とし
ての画質を3次元的に向上させることができる。さら
に、ウェーブレット変換に基づいて画像処理を行なって
いるので、所定信号の強調および雑音成分の低減を精度
良く行なうことができる。
As is apparent from the above description, according to the present invention, image processing is performed using all information in each dimension of a three-dimensional image. Can be improved three-dimensionally. Further, since the image processing is performed based on the wavelet transform, it is possible to enhance the predetermined signal and reduce the noise component with high accuracy.

【図面の簡単な説明】 [Brief description of the drawings]

【図1】実施例の3次元画像処理装置の概略構成を示すブロック図である。 FIG. 1 is a block diagram illustrating a schematic configuration of a three-dimensional image processing apparatus according to an embodiment.

【図2】被撮像対象を撮像して得られる複数の断層画像の様子を示す図である。 FIG. 2 is a diagram showing a state of a plurality of tomographic images obtained by imaging a subject to be imaged.

【図3】原画像メモリに記憶される3次元画像の様子を示す図である。 FIG. 3 is a diagram showing a state of a three-dimensional image stored in an original image memory.

【図4】ウェーブレット変換の簡単な説明を示す図である。 FIG. 4 is a diagram illustrating a brief description of a wavelet transform.

【図5】3次元画像の各方向にウェーブレット変換を施した様子を示す図である。 FIG. 5 is a diagram illustrating a state in which a wavelet transform is performed in each direction of a three-dimensional image.

【図6】被撮像対象および雑音成分の断面を示す図である。 FIG. 6 is a diagram illustrating a cross section of an object to be imaged and a noise component.

【図7】プロファイルラインでの濃度変化を示す画像信号の様子を示す図である。 FIG. 7 is a diagram illustrating a state of an image signal indicating a density change in a profile line.

【図8】プロファイルラインでの画像信号の詳細成分を示す図である。 FIG. 8 is a diagram showing detailed components of an image signal in a profile line.

【図9】強度変換関数を示す図である。 FIG. 9 is a diagram showing an intensity conversion function.

【図10】強度変換された詳細成分を示す図である。 FIG. 10 is a diagram showing detailed components subjected to intensity conversion.

【図11】変形例の強度変換関数を示す図である。 FIG. 11 is a diagram showing an intensity conversion function according to a modification.

【符号の説明】 [Explanation of symbols]

1 … 画像処理部 2 … 原画像メモリ 3 … 強度変換関数メモリ 4 … 入力部 5 … 表示部 11 … 多重解像度分解部 12 … 多重解像度成分メモリ 13 … 詳細成分強度変換部 14 … 画像再構成部 15 … 再構成画像メモリ M … 被撮像対象 K1〜Kn… 断層画像 DESCRIPTION OF SYMBOLS 1 ... Image processing part 2 ... Original image memory 3 ... Intensity conversion function memory 4 ... Input part 5 ... Display part 11 ... Multi-resolution decomposition part 12 ... Multi-resolution component memory 13 ... Detailed component intensity conversion part 14 ... Image reconstruction part 15 ... Reconstructed image memory M ... Object to be imaged K1 to Kn ... Tomographic image

───────────────────────────────────────────────────── フロントページの続き Fターム(参考) 4C093 AA26 BA07 CA01 CA06 EE01 FD04 FD20 FF09 FF15 FF42 FF46 FG05 FG07 FH02 4C096 AB01 AB07 DC11 DC18 DC36 DC37 DC40 DD01 DD09 FC16 5B057 AA09 CA02 CA08 CA13 CA16 CB02 CB13 CB16 CC02 CE02 CE03  ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) 4C093 AA26 BA07 CA01 CA06 EE01 FD04 FD20 FF09 FF15 FF42 FF46 FG05 FG07 FH02 4C096 AB01 AB07 DC11 DC18 DC36 DC37 DC40 DD01 DD09 FC16 5B057 AA09 CA02 CA08 CA13 CE02 CB02

Claims (1)

    【特許請求の範囲】 [Claims]
  1. 【請求項1】 被撮像対象の複数箇所で断層画像を撮像することで得られる3次元画像を記憶する画像記憶手段と、前記画像記憶手段に記憶された3次元画像の画像信号に画像処理を施す画像処理手段とを備える3次元画像処理装置において、前記画像処理手段は、(a)前記3
    次元画像の各次元方向の画像信号にウェーブレット変換を施すことにより、前記3次元画像を多重解像度成分に分解して、前記3次元画像の近似成分と複数の詳細成分とを求める多重解像度分解手段と、(b)前記多重解像度分解手段で求められる複数の詳細成分の中の任意の詳細成分に、予め設定された強度変換関数を作用させることで、前記任意の詳細成分に含まれる雑音成分を低減するとともに、所定成分を強調する強度変換を行なう強度変換手段と、(c)前記強度変換手段で強度変換された任意の詳細成分を含む複数の詳細成分と、前記近似成分とで構成される強度変換後多重解像度成分にウェーブレット逆変換を施すことにより、3次元画像を再構成する画像再構成手段とを備えたことを特徴とする3次元画像処理装置。 A multi-resolution decomposition means for decomposing the three-dimensional image into multi-resolution components by performing wavelet transform on the image signal in each dimension direction of the dimensional image to obtain an approximate component and a plurality of detailed components of the three-dimensional image. , (B) By applying a preset intensity conversion function to an arbitrary detailed component among a plurality of detailed components obtained by the multi-resolution decomposition means, the noise component contained in the arbitrary detailed component is reduced. In addition, the intensity is composed of an intensity conversion means that performs intensity conversion that emphasizes a predetermined component, (c) a plurality of detailed components including any detailed component that has been intensity-converted by the intensity conversion means, and the approximate component. A three-dimensional image processing apparatus including an image reconstructing means for reconstructing a three-dimensional image by performing an inverse wavelet transform on a multi-resolution component after conversion. An image storage unit configured to store a three-dimensional image obtained by capturing a tomographic image at a plurality of locations of an imaging target; and performing image processing on an image signal of the three-dimensional image stored in the image storage unit. A three-dimensional image processing apparatus comprising: An image storage unit configured to store a three-dimensional image obtained by capturing a tomographic image at a plurality of locations of an imaging target; and performing image processing on an image signal of the three-dimensional image stored in the image storage unit. three-dimensional image processing apparatus comprising:
    Multi-resolution decomposition means for decomposing the three-dimensional image into multi-resolution components by performing a wavelet transform on the image signal in each dimensional direction of the three-dimensional image to obtain an approximate component of the three-dimensional image and a plurality of detailed components; (B) reducing a noise component included in the arbitrary detailed component by applying a predetermined intensity conversion function to an arbitrary detailed component among a plurality of detailed components obtained by the multi-resolution decomposition means; And (c) a plurality of detailed components including an arbitrary detailed component subjected to the intensity conversion by the intensity converting unit, and an intensity constituted by the approximate component. A three-dimensional image processing apparatus, comprising: image reconstruction means for reconstructing a three-dimensional image by performing inverse wavelet transform on the conver Multi-resolution decomposition means for decomposing the three-dimensional image into multi-resolution components by performing a wavelet transform on the image signal in each dimensional direction of the three-dimensional image to obtain an approximate component of the three-dimensional image and a plurality of detailed components; (B) reducing a noise component included in the arbitrary detailed component by applying a predetermined intensity conversion function to an arbitrary detailed component among a plurality of detailed components obtained by the multi-resolution decomposition means; And (c) a plurality A three-dimensional image processing apparatus, comprising: image reconstruction means for reconstructing a three-dimensional image by performing. of detailed components including an arbitrary detailed component subjected to the intensity conversion by the intensity converting unit, and an intensity composed by the approximate component. inverse wavelet transform on the conver ted multi-resolution component. ted multi-resolution component.
JP10200754A 1998-07-15 1998-07-15 Three-dimensional image processor Pending JP2000030044A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP10200754A JP2000030044A (en) 1998-07-15 1998-07-15 Three-dimensional image processor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP10200754A JP2000030044A (en) 1998-07-15 1998-07-15 Three-dimensional image processor

Publications (1)

Publication Number Publication Date
JP2000030044A true JP2000030044A (en) 2000-01-28

Family

ID=16429630

Family Applications (1)

Application Number Title Priority Date Filing Date
JP10200754A Pending JP2000030044A (en) 1998-07-15 1998-07-15 Three-dimensional image processor

Country Status (1)

Country Link
JP (1) JP2000030044A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002043001A1 (en) * 2000-11-24 2002-05-30 Nihon University Image processing method
US9569820B2 (en) 2013-01-04 2017-02-14 Samsung Electronics Co., Ltd. Method and apparatus for image correction
US10110874B2 (en) 2013-05-28 2018-10-23 Toshiba Medical Systems Corporation Medical-image processing apparatus generating plural parallax images with different viewpoint positions based on adjusting parallactic angles

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002043001A1 (en) * 2000-11-24 2002-05-30 Nihon University Image processing method
US7068837B2 (en) 2000-11-24 2006-06-27 Nihon University Image processing method
US9569820B2 (en) 2013-01-04 2017-02-14 Samsung Electronics Co., Ltd. Method and apparatus for image correction
US10110874B2 (en) 2013-05-28 2018-10-23 Toshiba Medical Systems Corporation Medical-image processing apparatus generating plural parallax images with different viewpoint positions based on adjusting parallactic angles

Similar Documents

Publication Publication Date Title
Rueda et al. Single-image super-resolution of brain MR images using overcomplete dictionaries
Bhateja et al. Multimodal medical image sensor fusion framework using cascade of wavelet and contourlet transform domains
Ganasala et al. CT and MR image fusion scheme in nonsubsampled contourlet transform domain
Li et al. Performance comparison of different multi-resolution transforms for image fusion
JP5637928B2 (en) Medical image display device
Diwakar et al. A review on CT image noise and its denoising
US7760924B2 (en) System and method for generating a 2D image from a tomosynthesis data set
Wang et al. Detection of microcalcifications in digital mammograms using wavelets
Borsdorf et al. Wavelet based noise reduction in CT-images using correlation analysis
KR100574536B1 (en) Image processing device, image processing method, storage medium, and program
US6782137B1 (en) Digital image display improvement system and method
US8542944B2 (en) Method and apparatus for multi-scale based dynamic range compression and noise suppression
US6823090B2 (en) Image processing apparatus and its method, program, and storage medium
US7362845B2 (en) Method and apparatus of global de-noising for cone beam and fan beam CT imaging
US7489825B2 (en) Method and apparatus for creating a multi-resolution framework for improving medical imaging workflow
US7227980B2 (en) Systems and methods for tomographic reconstruction of images in compressed format
Stahl et al. Digital radiography enhancement by nonlinear multiscale processing
US7970203B2 (en) Purpose-driven data representation and usage for medical images
US6592523B2 (en) Computationally efficient noise reduction filter for enhancement of ultrasound images
US6246783B1 (en) Iterative filter framework for medical images
JP4891577B2 (en) Medical image display device
US6813335B2 (en) Image processing apparatus, image processing system, image processing method, program, and storage medium
Bhatnagar et al. Human visual system inspired multi-modal medical image fusion framework
US7916139B2 (en) Imaging volume data
US6463181B2 (en) Method for optimizing visual display of enhanced digital images