JP2001119586A5 - - Google Patents

Download PDF

Info

Publication number
JP2001119586A5
JP2001119586A5 JP1999293933A JP29393399A JP2001119586A5 JP 2001119586 A5 JP2001119586 A5 JP 2001119586A5 JP 1999293933 A JP1999293933 A JP 1999293933A JP 29393399 A JP29393399 A JP 29393399A JP 2001119586 A5 JP2001119586 A5 JP 2001119586A5
Authority
JP
Japan
Prior art keywords
image
multispectral
principal component
spectral
principal
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.)
Granted
Application number
JP1999293933A
Other languages
Japanese (ja)
Other versions
JP3986219B2 (en
JP2001119586A (en
Filing date
Publication date
Application filed filed Critical
Priority to JP29393399A priority Critical patent/JP3986219B2/en
Priority claimed from JP29393399A external-priority patent/JP3986219B2/en
Publication of JP2001119586A publication Critical patent/JP2001119586A/en
Publication of JP2001119586A5 publication Critical patent/JP2001119586A5/ja
Application granted granted Critical
Publication of JP3986219B2 publication Critical patent/JP3986219B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Description

【0002】
【従来の技術】
今日、デジタル画像処理の進歩によって、画像の色情報(明度、色相、彩度)を完全に表現する手段として、画像の各画素毎に分光情報(スペクトル画像)を備える画像、すなわちマルチスペクトル画像が利用されている。
このマルチスペクトル画像は、撮影被写体を、複数のバンド帯域に分割して各バンド帯域毎に撮影した複数のバンド画像から構成されるマルチバンド画像に基づいて分光反射率分布を各画像毎に推定して得られるものである。このマルチバンド画像は、赤(R)、緑(G)および青(B)画像からなる従来のRGBカラー画像では十分に表現できない色情報を再現することができ、例えばより正確な色再現の望まれる絵画の世界にとって有効である。そこで、この色情報を正確に再現するといった特徴を生かすために、例えば380〜780nmの撮影波長帯域を10nm帯域毎に区切って41バンドさらには5nm帯域毎に区切って81バンドといった多くのバンド数を備えたマルチバンド画像に基づいてマルチスペクトル画像を得ることが望まれる。
[0002]
[Prior Art]
Nowadays, with the progress of digital image processing, an image provided with spectral information (spectral image) for each pixel of an image, that is, a multispectral image, is a means for completely expressing color information (lightness, hue, saturation) of the image. It's being used.
In this multispectral image, a spectral reflectance distribution is estimated for each image based on a multiband image composed of a plurality of band images obtained by dividing the photographed subject into a plurality of band bands and photographing each band. It is obtained by This multi-band image can reproduce color information that can not be sufficiently represented by a conventional RGB color image consisting of red (R), green (G) and blue (B) images. For example, a more accurate color reproduction is desired Is effective for the world of paintings. Therefore, to take advantage of the feature of accurately reproducing this color information, for example, the imaging wavelength band of 380 to 780 nm is divided into 10 nm bands, 41 bands are further divided into 5 nm bands, and the number of bands is as large as 81 bands. It is desirable to obtain a multispectral image based on the provided multiband image.

このような問題に対して、マルチスペクトル画像の各画素ごとの分光情報から得られるスペクトル波形を3つの等色関数、例えばRGB表色形の等色関数で展開するとともに、等色関数で表されないスペクトル波形の部分を、主成分分析法を用いて、主成分基底ベクトルで展開し、その中からスペクトル画像の画像情報を代表する主成分を抽出して採用し、それ以外の主成分は取り除き、最終的に等色関数を含め合計6〜8個の基底ベクトルで上記スペクトル波形を表現する方法が提案されている(Th.Keusen,Multispectoral Color System wuth an Encoding Format Compatible with the Conventional Tristimulus Model,Journal of Imaging Science and Technology 40:510-515(1996))。これを用いて、上記スペクトル波形を6〜8個の基底ベクトルとそれに対応した係数の対とで表わすことによって、マルチスペクトル画像の画像データを圧縮することができる。特に、RGB表色形の等色関数で表される場合等色関数の係数は、R、GおよびBの三刺激値に対応するので、R、GおよびB画素による3刺激値に基づいて画像処理や画像表示等が行われる従来の画像処理装置や画像表示装置に対応して適合するように特別な変換を施す必要がなく、直接画像データを従来の画像処理装置や画像表示装置に送ることができるといった処理の低減に対して優れた効果を備える。 To solve such problems, spectral waveforms obtained from spectral information of each pixel of a multispectral image are expanded by three color matching functions, for example, color matching functions of RGB colorimetry, and are not represented by color matching functions. The part of the spectral waveform is expanded with a principal component basis vector using principal component analysis, from which the principal component representing the image information of the spectral image is extracted and adopted, and the other principal components are removed, Finally, a method for representing the above-mentioned spectral waveform with a total of 6 to 8 basis vectors including a color matching function has been proposed (Th. Keusen, Multispectral Color System wth an Encoding Format Compatible with the Conventional Tristimulus Model, Journal of Imaging Science and Technology 40: 510-515 (1996)). This can be used to compress image data of a multispectral image by representing the spectral waveform as 6 to 8 basis vectors and the corresponding coefficient pairs. In particular, since the coefficients of the color matching function in the case of being represented by the color matching function of the RGB colorimetric form correspond to the tristimulus values of R, G and B, based on the tristimulus values by R, G and B pixels There is no need to perform special conversion to be compatible with the conventional image processing apparatus or image display apparatus in which image processing, image display, etc. are performed, and direct image data is sent to the conventional image processing apparatus or image display apparatus. It has an excellent effect on the reduction of processing that can be done.

ここで、前記最適主成分数は、色空間上の測色値に基づいて決定されるのが好ましく、その際、最適主成分数は、前記主成分ベクトルと前記主成分画像の中から選ばれて構成される合成画像の測色値の画像情報の、前記マルチスペクトル画像に基づいて構成されるオリジナル画像の測色値の画像情報に対する誤差の値が、所定値以下となる最小の主成分数であり、あるいは、前記マルチスペクトル画像に対する寄与の大きい主成分ベクトルを、主成分ベクトルの寄与の大きい順に、順次含めて前記合成画像を求めた時の前記オリジナル画像に対する前記誤差の変動が、所定値以下に収まる最小の主成分数であるのが好ましい。 Here, the optimum number of main components is preferably determined based on the colorimetric value on the color space, wherein the optimum number of main components is selected from the main component vector and the main component image. The minimum number of main components of which the value of the error of the colorimetric information of the synthesized image with respect to the image information of the colorimetric value of the original image constructed based on the multispectral image is equal to or less than a predetermined value Or, the variation of the error with respect to the original image when the composite image is determined by sequentially including the principal component vector having the largest contribution to the multispectral image in the descending order of the contribution of the principal component vector is a predetermined value It is preferable that it is the minimum main component number which fits below.

図1は、本発明のマルチスペクトル画像の画像圧縮方法を実施し、本発明のマルチスペクトル画像の画像圧縮装置を含むマルチスペクトル画像取得システム(以下、本システムという)10を示す。
本システム10は、撮影被写体Oを撮影し、得られたマルチスペクトル画像MS の画像データを記録メディアに保存するものであって、撮影被写体Oを照らす光源12と、撮影波長帯域を複数のバンド帯域に分割する可変フィルタ14と、撮影被写体Oを撮影してマルチバンド画像MBを得るCCDカメラ16と、画像データを一時保持するマルチバンド画像記憶装置18と、マルチバンド画像から各画素毎に分光反射率分布を推定してマルチスペクトル画像MSを得るマルチスペクトル画像取得装置20と、マルチスペクトル画像MS の画像データを、視覚的な劣化が少なく、圧縮率を高くして圧縮するマルチスペクトル画像圧縮装置22と、得られた圧縮画像データを保存する記録メディアドライブ装置24とを主に有して構成される。なお、本発明において、マルチスペクトル画像Msは、少なくとも6チャンネル以上のスペクトル画像を備え、すなわち、分光反射率分布のデータを持つ構成波長数が6以上であるのが好ましい。
FIG. 1 shows a multispectral image acquisition system (hereinafter referred to as the present system) 10 that implements the multispectral image image compression method of the present invention and includes the multispectral image image compression device of the present invention.
The present system 10 shoots a photographic subject O and stores the obtained image data of the multispectral image MS in a recording medium, and the light source 12 for illuminating the photographic subject O, a plurality of imaging wavelength bands in a plurality of bands It is divided into a variable filter 14, the spectral and CCD camera 16 to obtain a multiband image M B by photographing the photographic subject O, a multi-band image storage unit 18 for storing image data temporarily from the multi-band image for each pixel a multispectral image acquisition device 20 to obtain a multispectral image M S to estimate the reflectance distribution, the image data of the multispectral image MS, less visual deterioration, multispectral image compression to increase to compress the compression ratio It mainly comprises an apparatus 22 and a recording media drive apparatus 24 for storing the obtained compressed image data. In the present invention, it is preferable that the multispectral image M s comprises spectral images of at least six channels or more, that is, the number of constituent wavelengths having spectral reflectance distribution data is six or more.

主成分分析部22aは、マルチスペクトル画像MSの各画素毎に備える分光反射率分布の主成分分析を行い、各画素ごとに分光反射率分布を主成分に展開する部分である。なお、以降では、撮影波長帯域を複数のバンド帯域に分割するバンド数をnとして説明する。
本発明における主成分分析法は、観測波形データ群を正規直交展開して標本化する方法の一つで、最適標本化ともいわれるものである。即ち、最も少ない数の直交基底関数の加重平均で、観測波形データを最も精度良く表現するための方法である。ここでは、直交基底関数を主成分ベクトルと呼ぶ。
本発明における主成分分析として具体的には、観測波形から、統計的手法および固有値解析法を用いて、観測波形に固有の1次独立な固有ベクトルを主成分ベクトルとして求め、この主成分ベクトルから、本来観測波形に雑音成分が無ければ、固有値が0となる固有値の小さな主成分ベクトルを取り除き、バンド数nより少ない数の最適主成分ベクトルを求め、この最適主成分ベクトルによって観測波形を線型的に表す、南茂夫著、「科学計測のための波形データ処理」、220−225頁に記載の方法があげられる。この分析方法は、主成分分析部22aおよび後述する最適主成分ベクトル・画像抽出部22bにおいて主に行われる。
主成分分析法を用いる場合には、観測波形であるマルチスペクトル画像MSの画素毎の分光反射率波形が、線型的に表すことができ、また分光反射率波形に含まれる雑音成分も、分光反射率の値と無関係な雑音であることが好ましい。
Principal component analysis unit 22a performs a principal component analysis of spectral reflectance distribution provided for each pixel of the multi-spectral image M S, is a portion for deploying the spectral reflectance distribution for each pixel in the main component. In the following, the number of bands for dividing the imaging wavelength band into a plurality of band bands will be described as n.
The principal component analysis method according to the present invention is one of the methods of performing orthonormal expansion and sampling of the observation waveform data group, and is also called optimum sampling. That is, it is a method for representing the observed waveform data with the highest accuracy by the weighted average of the least number of orthogonal basis functions. Here, the orthogonal basis function is called a principal component vector.
Specifically, as the principal component analysis in the present invention, a linear independent eigenvector unique to the observation waveform is determined as a principal component vector from the observation waveform using a statistical method and an eigen value analysis method, and from the principal component vector If the observed waveform has no noise component, the principal component vector with a small eigenvalue with an eigenvalue of 0 is removed, the number of optimum principal component vectors smaller than the number of bands n is determined, and the observation waveform is linearized by this optimum principal component vector South Shigeo, "Processing of waveform data for scientific measurement", pp. 220-225. This analysis method is mainly performed in the principal component analysis unit 22a and the optimum principal component vector / image extraction unit 22b described later.
When using the principal component analysis method, spectral reflectance waveform for each pixel of an observed waveform multispectral image M S is, it can be expressed linearly, also the noise component included in the spectral reflectance waveforms, spectral Preferably, the noise is irrelevant to the reflectance value.

JPEG方式とは、例えば1024×1024画素の主成分画像Skを8×8画素のブロック画像に分解し、このブロック画像各々に対して、cosine関数による次元の離散型のフーリエ展開であるDCTを施し、得られ低周波成分から高周波成分に至る複数のフーリエ係数をDCT係数として求めたのち、予め与えられた量子化テーブルによって上記DCT係数を除して、高周波成分のフーリエ係数を0として省略することで、高周波成分の画像データを圧縮し、その後DCT係数の0次低周波成分である直流成分とそれ以外の周波数成分に分け、ハフマン符号化方式や公知の算術符号化方式を用いて、DCT係数の画像データを符号化し圧縮する方式である。ここで、上記量子化テーブルの値は、主成分画像Skの像構造によって変化するものである。
本発明においては、上記DCT係数の高周波成分を量子化テーブルによって除去した画像データを、ハフマン符号化方式や公知の算術符号化方式を用いることなく、圧縮マルチスペクトル画像データとして、画像圧縮部22cから出力させてもよい。また、最適主成分画像Sk(k=1〜m1)の画像像データに対して、符号化による圧縮を直接施してもよい。
In the JPEG method, for example, a principal component image S k of 1024 × 1024 pixels is decomposed into a block image of 8 × 8 pixels, and a DCT, which is a two- dimensional discrete Fourier expansion by a cosine function, is generated for each block image. And the plurality of Fourier coefficients from the low frequency component to the high frequency component are obtained as DCT coefficients, and then the DCT coefficient is divided by a predetermined quantization table to set the Fourier coefficient of the high frequency component to 0. By omitting it, the image data of high frequency component is compressed, and then it is divided into the direct current component which is the 0th low frequency component of the DCT coefficient and the other frequency component, and using Huffman coding method or known arithmetic coding method. , A method of encoding and compressing image data of DCT coefficients. Here, the value of the quantization table is used varies depending image structure of the main component image S k.
In the present invention, the image data from which the high-frequency components of the DCT coefficient are removed by the quantization table is not compressed using the Huffman coding method or the well-known arithmetic coding method, but is used as compressed multispectral image data from the image compression unit 22c. You may output it. Further, compression by encoding may be directly applied to the image data of the optimum main component image S k (k = 1 to m 1 ).

まず、光源12、可変フィルタ14およびCCDカメラ16によって形成されるマルチバンドカメラによって撮影被写体Oを撮影し、複数のバンド帯域に分割された複数のバンド画像からなるマルチバンド画像MBを取得する(ステップ100)。得られたマルチバンド画像MBは、マルチバンド画像データ記憶装置18に一時記憶されると共に、マルチスペクトル画像取得装置20に送られる。 First, the light source 12, the multi-band camera which is formed by the variable filter 14 and the CCD camera 16 to shoot the photographic subject of O, acquires multiband image M B consisting of a plurality of bands image divided into a plurality of-band ( Step 100). The resulting multi-band image M B is a multi-band image data storage device 18 to both the stored temporarily Ru, it is sent to the multispectral image acquisition device 20.

Claims (1)

前記最適主成分数は、前記主成分ベクトルと前記主成分画像の中から選ばれて構成される合成画像の測色値の画像情報の、前記マルチスペクトル画像に基づいて構成されるオリジナル画像の測色値の画像情報に対する誤差の値が、所定値以下となる最小の主成分数であり、あるいは、前記マルチスペクトル画像に対する寄与の大きい主成分ベクトルを、主成分ベクトルの寄与の大きい順に、順次含めて前記合成画像を求めた時の前記オリジナル画像に対する前記誤差の変動が、所定値以下に収まる最小の主成分数である請求項1または2に記載のマルチスペクトル画像の画像圧縮方法。The optimal number of main components is a measure of an original image configured based on the multispectral image of image information of colorimetric values of a composite image configured by selecting from among the main component vector and the main component image. It is the minimum number of principal components in which the value of the error with respect to the image information of the color value is equal to or less than a predetermined value, or the principal component vectors having the largest contribution to the multispectral image are sequentially included in the descending order of the contribution of the principal components The method according to claim 1 or 2, wherein the variation of the error with respect to the original image when the synthesized image is obtained is the minimum number of main components falling within a predetermined value or less.
JP29393399A 1999-10-15 1999-10-15 Image compression method and image compression apparatus for multispectral image Expired - Fee Related JP3986219B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP29393399A JP3986219B2 (en) 1999-10-15 1999-10-15 Image compression method and image compression apparatus for multispectral image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP29393399A JP3986219B2 (en) 1999-10-15 1999-10-15 Image compression method and image compression apparatus for multispectral image

Publications (3)

Publication Number Publication Date
JP2001119586A JP2001119586A (en) 2001-04-27
JP2001119586A5 true JP2001119586A5 (en) 2005-05-26
JP3986219B2 JP3986219B2 (en) 2007-10-03

Family

ID=17801055

Family Applications (1)

Application Number Title Priority Date Filing Date
JP29393399A Expired - Fee Related JP3986219B2 (en) 1999-10-15 1999-10-15 Image compression method and image compression apparatus for multispectral image

Country Status (1)

Country Link
JP (1) JP3986219B2 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002098116A1 (en) * 2001-05-28 2002-12-05 Odagawa, Kazuo Voice information providing system and voice information providing method
JP5727109B2 (en) 2011-12-28 2015-06-03 ドルビー ラボラトリーズ ライセンシング コーポレイション Spectral image processing
US9754389B2 (en) 2012-08-06 2017-09-05 Koninklijke Philips N.V. Image noise reduction and/or image resolution improvement
CN114630054A (en) * 2020-12-14 2022-06-14 瑞昱半导体股份有限公司 Image acquisition device with image flicker detection mechanism and image flicker detection method thereof

Similar Documents

Publication Publication Date Title
JP4097873B2 (en) Image compression method and image compression apparatus for multispectral image
JP4884734B2 (en) System and method for encoding mosaic image data using reversible color conversion
US5053861A (en) Compression method and apparatus for single-sensor color imaging systems
US5065229A (en) Compression method and apparatus for single-sensor color imaging systems
US20170339430A1 (en) Encoder, decoder, method of encoding data, method of decoding data, utilizing data format transformations
Malvar et al. Progressive-to-lossless compression of color-filter-array images using macropixel spectral-spatial transformation
US6819801B2 (en) System and method for processing demosaiced images to reduce color aliasing artifacts
JP3986221B2 (en) Image compression method and image compression apparatus for multispectral image
JP2001119587A5 (en)
JP4097874B2 (en) Image compression method and image compression apparatus for multispectral image
WO2014018050A1 (en) Techniques for Video Compression
US7327876B2 (en) Image processing device
Lukac et al. Single-sensor camera image compression
Lee et al. Lossless compression of HDR color filter array image for the digital camera pipeline
JP2001119586A5 (en)
US7146055B2 (en) Image processing decompression apparatus and method of using same different scaling algorithms simultaneously
JP2001160904A (en) Image processor, image processing system, image-coding method and storage medium
Konig et al. Multispectral image encoding
JP3986219B2 (en) Image compression method and image compression apparatus for multispectral image
JPH1169377A (en) Image information encoding device and device applying the same
Shinoda et al. Multispectral image compression for spectral and color reproduction based on lossy to lossless coding
RU2295839C2 (en) Method for compressing digital images and device for realization of said method
Cuce et al. Compression of images in CFA format
JP2004159045A (en) Method, apparatus and program for encoding multi-band video signal and recording medium with the program recorded thereon
JPH0797829B2 (en) Color image input / output device