WO2019047025A1 - Method for forming local feature descriptor of hyperspectral image and forming system - Google Patents

Method for forming local feature descriptor of hyperspectral image and forming system Download PDF

Info

Publication number
WO2019047025A1
WO2019047025A1 PCT/CN2017/100581 CN2017100581W WO2019047025A1 WO 2019047025 A1 WO2019047025 A1 WO 2019047025A1 CN 2017100581 W CN2017100581 W CN 2017100581W WO 2019047025 A1 WO2019047025 A1 WO 2019047025A1
Authority
WO
WIPO (PCT)
Prior art keywords
vector
neighborhood
length
feature
descriptor
Prior art date
Application number
PCT/CN2017/100581
Other languages
French (fr)
Chinese (zh)
Inventor
李岩山
唐浩劲
谢维信
Original Assignee
深圳大学
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 深圳大学 filed Critical 深圳大学
Priority to PCT/CN2017/100581 priority Critical patent/WO2019047025A1/en
Publication of WO2019047025A1 publication Critical patent/WO2019047025A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features

Definitions

  • the present invention relates to the field of image processing technologies, and in particular, to a method and a system for forming a local feature descriptor of a hyperspectral image.
  • This method of combining spatial information and spectral information is generally applied to face recognition.
  • Pan et al. pioneered a new method of hyperspectral face recognition by extracting spectral features on features such as forehead, cheeks, and lips, but this method did not apply to any spatial information.
  • Unzir et al. developed an aerial spectrum special authentication extraction method to calculate the low frequency factor of hyperspectral face images.
  • Hyperspectral imaging offers new opportunities for interpersonal facial recognition, however, the extraction of compact and discriminative features from high-dimensional hyperspectral image cubes is a challenging task.
  • the 3D discrete cosine transform optimally compresses the information in the low frequency coefficients, represents each hyperspectral facial cube with a small number of low frequency DCT coefficients, and develops a partial least squares (PLS) regression for accurate classification.
  • PLS partial least squares
  • a spatial spectral feature descriptor is then generated by applying a 3D histogram on the derivative pattern, which can be used to convert the hyperspectral facial image to a vectorized representation.
  • this method can describe unique microscopic patterns that integrate potential spatial and spectral information.
  • Hyperspectral images and ordinary two-dimensional images have many different characteristics. They add spectral dimensions based on two-dimensional image information to form a three-dimensional coordinate space. If each band data of the hyperspectral image is regarded as a layer, and the imaging spectrum data is expressed as a whole in the coordinate space, a three-dimensional data cube having multiple layers formed by overlapping in a band sequence is formed. Hyperspectral images have many bands, which can provide hundreds or even thousands of bands per pixel. These bands are generally less than 10 cm in length, and the bands are continuous. Some sensors can provide near-solar spectrum. Continuous ground spectrum.
  • hyperspectral data and the information it carries are generally represented by three spatial representations: image space with spatial geometric positional relationship, spectral space containing rich spectral information, and suitable for patterns.
  • image space with spatial geometric positional relationship image space with spatial geometric positional relationship
  • spectral space containing rich spectral information spectral space containing rich spectral information
  • suitable for patterns The feature space of the application in recognition.
  • Hyperspectral remote sensing techniques can combine spectra that determine the properties of a substance or feature with images that reveal their spatial and geometric relationships. The corresponding spectral curves of different materials are also different.
  • an object of the present invention is to provide a method and a system for forming a local feature descriptor of a hyperspectral image, which aims to solve the problem that the local feature description algorithm of the two-dimensional image in the prior art is not applicable to the hyperspectral image. .
  • the invention provides a method for forming a local feature descriptor of a hyperspectral image, wherein the method comprises:
  • Feature description step using the angle between the spectral vectors and the modulus of the spectral vector to describe the neighborhood features;
  • Forming a description sub-step forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
  • the step of describing the feature specifically includes:
  • the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
  • the angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
  • the step of describing the feature specifically includes:
  • the angle between the neighborhood vector vector p and the first dimension vector of the neighborhood vector vector p is synthesized.
  • the step of describing the feature specifically includes:
  • the first value describing the neighborhood spatial feature is obtained from the original LBP operator expression. among them,
  • the second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • the third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • the forming the sub-step includes:
  • the three descriptors Descriptor norm , A join is made to obtain a new descriptor of length 3*2 ⁇ P, the length of the new descriptor being three times the length of the original local binary pattern vector.
  • the present invention also provides a system for forming a local feature descriptor of a hyperspectral image, the system comprising:
  • a feature description module for describing a neighborhood feature by using an angle between spectral vectors and a spectral vector
  • a description sub-module is formed for forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
  • the feature description module is specifically configured to:
  • the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
  • the angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
  • the feature description module is further used to:
  • Is the angle between vector values of the intermediate vector p neighborhood of the center vector of the synthesis of one-dimensional vector c is a vector, wherein the vector center vector c local neighborhood of V, neighborhood radius is R, the neighborhood points is P, o
  • the angle between the neighborhood vector vector p and the first dimension vector of the neighborhood vector vector p is synthesized.
  • the feature description module is further used to:
  • the first value describing the neighborhood spatial feature is obtained from the original LBP operator expression. among them,
  • the second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • the third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • the forming description submodule is specifically configured to:
  • the three descriptors Descriptor norm , A join is made to obtain a new descriptor of length 3*2 ⁇ P, the length of the new descriptor being three times the length of the original local binary pattern vector.
  • the technical solution provided by the invention utilizes the angle between the spectral vectors and the characterization of the neighborhood features of the spectral vector to form a vector whose length is three times the length of the original local binary pattern vector.
  • the narration combined with the spatial position information and spectral vector information of the neighborhood of hyperspectral feature points, can be applied to hyperspectral images.
  • FIG. 1 is a flow chart showing a method for forming a local feature descriptor of a hyperspectral image according to an embodiment of the present invention
  • FIG. 2 is a neighborhood representation of a central spectral vector in accordance with an embodiment of the present invention.
  • FIG. 3 is a schematic diagram showing the internal structure of a forming system 10 for a local feature descriptor of a hyperspectral image according to an embodiment of the present invention.
  • a specific embodiment of the present invention provides a method for forming a local feature descriptor of a hyperspectral image, wherein the method mainly includes the following steps:
  • Feature description step using the angle between the spectral vectors and the modulus of the spectral vector to describe the neighborhood features;
  • Forming a description sub-step forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
  • the invention provides a method for forming a local feature descriptor of a hyperspectral image, which uses the angle between the spectral vectors and the model of the spectral vector to describe the feature of the neighborhood feature to form a vector length which is the original local binary mode (Local Binary Patterns (LBP) triple the descriptor of the vector length, combined with the spatial position information and spectral vector information of the hyperspectral feature point neighborhood, can be applied to hyperspectral images.
  • LBP Local Binary Patterns
  • FIG. 1 is a flowchart of a method for forming a local feature descriptor of a hyperspectral image according to an embodiment of the present invention.
  • step S1 the feature description step describes the neighborhood feature by using the angle between the spectral vectors and the mode of the spectral vector.
  • the feature description step specifically includes:
  • the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
  • the angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
  • the hyperspectral image contains another aspect of information, that is, the spectral response of the object, since the spectral vector of the hyperspectral image contains rich feature information. Therefore, it is more efficient to replace the gray value with the spectral vector.
  • FIG. 2 draws a neighborhood representation of the central spectral vector, where the line indicated by 201 is the central spectral vector.
  • FIG. 2 there is shown a neighborhood representation of a central spectral vector in accordance with an embodiment of the present invention.
  • the center of the cube and the arrow to the right represent the central spectral vector
  • a feature point in the spectral domain of the hyperspectral image is selected as the central pixel
  • the DN value of the central pixel x c , y c , ⁇ c
  • the upper DN value constitutes a set of vectors v c
  • the DN value of the neighborhood cell of the local neighborhood can be composed of a series of vectors v p .
  • the feature description step specifically includes:
  • the length of the neighborhood vector vector p is the same as the center vector vector c ;
  • the feature description step specifically includes:
  • the first value describing the neighborhood spatial feature is obtained from the original LBP operator expression. among them,
  • the second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • the third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • step S2 a description sub-step is formed: a descriptor whose vector length is three times the length of the original local binary pattern vector is formed according to the description result.
  • the forming the sub-steps specifically includes:
  • the three descriptors Descriptor norm , A join is made to obtain a new descriptor of length 3*2 ⁇ P, the length of the new descriptor being three times the length of the original local binary pattern vector.
  • the hyperspectral image organically combines the spectral information reflecting the radiation property of the substance with the two-dimensional image information reflecting the geometric relationship of the object space, so that the hyperspectral image can be grayscaled.
  • the image and the color image provide more information.
  • the image of the "integration of the map" proposed by the present invention combines the advantages of the two-dimensional image and the spectral information, and broadens the analysis method of the image, which is very important for image analysis and recognition. The meaning.
  • the invention provides a method for forming a local feature descriptor of a hyperspectral image, and proposes a novel descriptor based on a local binary pattern for the feature of the spectral vector of the hyperspectral image.
  • the angle between the spectral vectors and the model of the spectral vector are used to describe the features of the neighborhood, and a descriptor whose vector length is three times the length of the original local binary pattern vector is formed.
  • the algorithm takes into account the characteristics of the hyperspectral image and combines the spatial position information and spectral vector information of the neighborhood of the hyperspectral feature points.
  • a specific embodiment of the present invention further provides a system for forming a local feature descriptor of a hyperspectral image, which mainly includes:
  • a feature description module for describing a neighborhood feature by using an angle between spectral vectors and a spectral vector
  • a description sub-module is formed for forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
  • the present invention provides a hyperspectral image local feature descriptor forming system 10, which utilizes the angle between the spectral vectors and the spectral vector to model the feature description of the neighborhood feature to form a vector length which is the original local binary pattern vector.
  • the descriptor of three times the length, combined with the spatial position information and spectral vector information of the neighborhood of the hyperspectral feature points, can be applied to hyperspectral images.
  • FIG. 3 a schematic structural diagram of a system 10 for forming a local feature descriptor of a hyperspectral image according to an embodiment of the present invention is shown.
  • the hyperspectral image local feature descriptor forming system 10 mainly includes a feature description module 11 and a formation description sub-module 12.
  • the feature description module 11 is configured to describe the neighborhood features by using the angle between the spectral vectors and the modulus of the spectral vector.
  • the feature description module 11 is specifically configured to:
  • the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
  • the angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
  • the feature description module 11 is further specifically configured to:
  • the length of the neighborhood vector vector p is the same as the center vector vector c ;
  • the feature description module 11 is further specifically configured to:
  • the first value describing the neighborhood spatial feature is obtained from the original LBP operator expression. among them,
  • the second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • the third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
  • Forming a description sub-module 12 for forming a vector length according to the description result is an original local binary value A three-dimensional descriptor of the length of the pattern vector.
  • the forming description sub-module 12 is specifically configured to:
  • the three descriptors Descriptor norm , A join is made to obtain a new descriptor of length 3*2 ⁇ P, the length of the new descriptor being three times the length of the original local binary pattern vector.
  • the present invention provides a hyperspectral image local feature descriptor forming system 10, which utilizes the angle between the spectral vectors and the spectral vector to model the feature description of the neighborhood feature to form a vector length which is the original local binary pattern vector.
  • the descriptor of three times the length, combined with the spatial position information and spectral vector information of the neighborhood of the hyperspectral feature points, can be applied to hyperspectral images.
  • each unit included is only divided according to functional logic, but is not limited to the above division, as long as the corresponding function can be implemented; in addition, the specific name of each functional unit is also They are only used to facilitate mutual differentiation and are not intended to limit the scope of the present invention.

Abstract

A method for forming a local feature descriptor of a hyperspectral image, the method comprising: a feature description step: describing a neighborhood feature by using an included angle between spectral vectors and a norm of the spectrum vectors (S1); a descriptor formation step: according to a description result, forming a descriptor having a vector length that is three times the length of an original local binary mode vector (S2). Further provided is a system for forming a local feature descriptor of a hyperspectral image: forming a descriptor having a vector length that is three times the length of an original local binary mode vector by using the feature description capability of an included angle between spectrum vectors and a norm of the spectrum vectors for a neighborhood feature, while combining spatial position information and spectral information of a neighborhood of a hyperspectral feature point. The system may be applied to hyperspectral images.

Description

高光谱图像局部特征描述子的形成方法及形成系统Method for forming local feature descriptors of hyperspectral image and forming system 技术领域Technical field
本发明涉及图像处理技术领域，尤其涉及一种高光谱图像局部特征描述子的形成方法及形成系统。The present invention relates to the field of image processing technologies, and in particular, to a method and a system for forming a local feature descriptor of a hyperspectral image.
背景技术Background technique
结合空间信息和光谱信息的这种方法一般应用在人脸识别上,结合空谱特征主要有两方面难题，第一个是现有的高光谱人脸图像的质量很低，第二个就是空间信息和光谱信息的固有性质完全不同。为了解决这些难题，有许多不同的方法被学者们提出。Pan等通过在例如前额、脸颊、嘴唇等这些特征区域上提取光谱特征而开创了高光谱人脸识别的新方法，但是这种方法没有应用到任何空间信息。Unzir等在3D离散余弦转换的基础上发展了一种空谱特认证提取方法来计算高光谱人脸图像的低频因数。高光谱成像为人际面部鉴别提供了新的机会，然而，从高维度高光谱图像立方体的紧凑和辨别特征提取是一项具有挑战性的任务。3D离散余弦转换最佳地压缩了低频系数中的信息，用少量低频DCT系数表示每个高光谱面部立方体，并且制定了用于准确分类的部分最小二乘法(PLS)回归。但是这些方法要不就是只运用了光谱特征，要不就是把高光谱数据当作各向同性体。所以Jie等在三维局部微分模式的基础上提出了一种三维高阶纹理模式描述子。它为构建多方向和多邻域的局部微分模式提供一个框架，这种模式同时结合空间信息和光谱信息并且减少了高光谱图像中噪声的负面影响。传统2D人脸识别已经研究多年，取得了巨大的成功。尽管如此，除了面部空间域中的结构和纹理之外，还需要探索除外的信息。高光谱成像通过提供关于对象的附加光谱信息来满足这样的要求，完成了2D图像中提取的传统空间特征。基于局部导数模型，用空间光谱空间中的多方向导数和二值化 函数对高光谱面进行编码。然后，通过在导数图案上应用3D直方图来生成空间光谱特征描述符，其可以用于将高光谱面部图像转换为向量化表示。与传统的脸部识别方法相比，这种方法能够描述将潜在的空间和光谱信息整合在一起的独特微观图案。This method of combining spatial information and spectral information is generally applied to face recognition. There are two main problems in combining spatial spectrum features. The first is that the quality of existing hyperspectral face images is very low, and the second is space. The inherent properties of information and spectral information are quite different. In order to solve these problems, there are many different methods proposed by scholars. Pan et al. pioneered a new method of hyperspectral face recognition by extracting spectral features on features such as forehead, cheeks, and lips, but this method did not apply to any spatial information. Based on the 3D discrete cosine transform, Unzir et al. developed an aerial spectrum special authentication extraction method to calculate the low frequency factor of hyperspectral face images. Hyperspectral imaging offers new opportunities for interpersonal facial recognition, however, the extraction of compact and discriminative features from high-dimensional hyperspectral image cubes is a challenging task. The 3D discrete cosine transform optimally compresses the information in the low frequency coefficients, represents each hyperspectral facial cube with a small number of low frequency DCT coefficients, and develops a partial least squares (PLS) regression for accurate classification. But these methods use only spectral features, or hyperspectral data as isotropic. Therefore, Jie et al. proposed a three-dimensional high-order texture pattern descriptor based on the three-dimensional local differential mode. It provides a framework for constructing multi-directional and multi-neighbor local differential modes that combine spatial information and spectral information and reduce the negative effects of noise in hyperspectral images. Traditional 2D face recognition has been studied for many years and has achieved great success. Nevertheless, in addition to the structure and texture in the face space domain, it is necessary to explore the exclusion information. Hyperspectral imaging satisfies such requirements by providing additional spectral information about the object, completing the traditional spatial features extracted in the 2D image. Multi-directional derivative and binarization in spatial spectral space based on local derivative model The function encodes the hyperspectral surface. A spatial spectral feature descriptor is then generated by applying a 3D histogram on the derivative pattern, which can be used to convert the hyperspectral facial image to a vectorized representation. Compared to traditional face recognition methods, this method can describe unique microscopic patterns that integrate potential spatial and spectral information.
高光谱图像和普通的二维图像有许多不同的特点，它在二维图像信息的基础上添加光谱维，从而形成三维的坐标空间。如果把高光谱图像的每个波段数据都看成是一个层面，将成像光谱数据整体表达到该坐标空间，就会形成一个按波段顺序叠合构成的拥有多个层面的三维数据立方体。高光谱图像的波段比较多，它可以为每个像素提供数百甚至上千个波段，这些波段的范围一般小于10厘米，而且波段是连续的，有些传感器可以在某些太阳光谱范围内提供近乎连续的地物光谱。根据高光谱图像的特点及其相关处理技术的需要，高光谱数据和其携带的信息一般由三种空间表达方式：有空间几何位置关系的图像空间、包含丰富光谱信息的光谱空间以及适合于模式识别中的应用的特征空间。高光谱的遥感技术可以把确定物质或地物性质的光谱与揭示其空间和几何关系的图像结合在一起，不同的物质其对应的光谱曲线也是不一样的。Hyperspectral images and ordinary two-dimensional images have many different characteristics. They add spectral dimensions based on two-dimensional image information to form a three-dimensional coordinate space. If each band data of the hyperspectral image is regarded as a layer, and the imaging spectrum data is expressed as a whole in the coordinate space, a three-dimensional data cube having multiple layers formed by overlapping in a band sequence is formed. Hyperspectral images have many bands, which can provide hundreds or even thousands of bands per pixel. These bands are generally less than 10 cm in length, and the bands are continuous. Some sensors can provide near-solar spectrum. Continuous ground spectrum. According to the characteristics of hyperspectral images and the needs of related processing techniques, hyperspectral data and the information it carries are generally represented by three spatial representations: image space with spatial geometric positional relationship, spectral space containing rich spectral information, and suitable for patterns. The feature space of the application in recognition. Hyperspectral remote sensing techniques can combine spectra that determine the properties of a substance or feature with images that reveal their spatial and geometric relationships. The corresponding spectral curves of different materials are also different.
因此，虽然二维图像的局部特征描述算法研究已经非常成熟了，但是对于高光谱图像，除了包含空间信息以外，还包含了光谱信息，目前的二维图像的局部特征描述方法已经不适用于高光谱图像，所以亟需提供一种能适用于高光谱图像的局部特征描述算法。Therefore, although the local feature description algorithm of two-dimensional images has been very mature, for hyperspectral images, in addition to spatial information, spectral information is included. The current local feature description method of two-dimensional images is not suitable for high Spectral images, so there is a need to provide a local characterization algorithm that can be applied to hyperspectral images.
发明内容Summary of the invention
有鉴于此，本发明的目的在于提供一种高光谱图像局部特征描述子的形成方法及形成系统，旨在解决现有技术中的二维图像的局部特征描述算法不适用于高光谱图像的问题。In view of this, an object of the present invention is to provide a method and a system for forming a local feature descriptor of a hyperspectral image, which aims to solve the problem that the local feature description algorithm of the two-dimensional image in the prior art is not applicable to the hyperspectral image. .
本发明提出一种高光谱图像局部特征描述子的形成方法，其中，所述方法包括： The invention provides a method for forming a local feature descriptor of a hyperspectral image, wherein the method comprises:
特征描述步骤：利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述；Feature description step: using the angle between the spectral vectors and the modulus of the spectral vector to describe the neighborhood features;
形成描述子步骤：根据描述结果形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。Forming a description sub-step: forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
优选的，所述特征描述步骤具体包括：Preferably, the step of describing the feature specifically includes:
定义
Figure PCTCN2017100581-appb-000001
为中心矢量vectorc与合成所述中心矢量vectorc的中间一维矢量的夹角值，其中，中心矢量vectorc的长度为N，中心矢量vectorc的模为normc，N为奇数，half为长度N的中间值；definition
Figure PCTCN2017100581-appb-000001
Is the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
定义
Figure PCTCN2017100581-appb-000002
为中心矢量vectorc与合成所述中心矢量vectorc的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000002
The angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
优选的，所述特征描述步骤具体还包括：Preferably, the step of describing the feature specifically includes:
定义
Figure PCTCN2017100581-appb-000003
为邻域矢量vectorp与合成所述邻域矢量vectorp的中间一维矢量的夹角值，其中，中心矢量vectorc的局部邻域为V，邻域半径为R，邻域点数为P，邻域矢量vectorp(p＝0,1,...,P-1)的模为normp，且邻域矢量vectorp的长度与中心矢量vectorc相同；definition
Figure PCTCN2017100581-appb-000003
Vector for the neighborhood in the synthesis of the vector p o value of the intermediate field vector angle of a vector p-dimensional vector, wherein the vector center vector c local neighborhood of V, neighborhood radius is R, the neighborhood points is P, The modulus of the neighborhood vector vector p (p=0, 1, . . . , P-1) is norm p , and the length of the neighborhood vector vector p is the same as the center vector vector c ;
定义
Figure PCTCN2017100581-appb-000004
为邻域矢量vectorp与合成所述邻域矢量vectorp的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000004
The angle between the neighborhood vector vector p and the first dimension vector of the neighborhood vector vector p is synthesized.
优选的，所述特征描述步骤具体还包括：Preferably, the step of describing the feature specifically includes:
由原始LBP算子表达式得到描述邻域空间特征的第一个数值
Figure PCTCN2017100581-appb-000005
其中，
Figure PCTCN2017100581-appb-000006
The first value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000005
among them,
Figure PCTCN2017100581-appb-000006
由原始LBP算子表达式得到描述邻域空间特征的第二个数值
Figure PCTCN2017100581-appb-000007
The second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000007
由原始LBP算子表达式得到描述邻域空间特征的第三个数值
Figure PCTCN2017100581-appb-000008
The third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000008
优选的，所述形成描述子步骤具体包括：Preferably, the forming the sub-step includes:
分别将中心矢量vectorc和邻域矢量vectorp(p＝0,1,...,P-1)的ARC0、norm、ARC1进行比较，以得到长度为2^P三种描述子
Figure PCTCN2017100581-appb-000009
descriptornorm、
Figure PCTCN2017100581-appb-000010
Comparing the central vector vector c with the ARC 0 , norm , and ARC 1 of the neighborhood vector vector p (p=0, 1, . . . , P-1), respectively, to obtain three descriptors of length 2^P.
Figure PCTCN2017100581-appb-000009
Descriptor norm ,
Figure PCTCN2017100581-appb-000010
将所述三种描述子
Figure PCTCN2017100581-appb-000011
descriptornorm、
Figure PCTCN2017100581-appb-000012
进行联结，以得到长度为3*2^P的新描述子，所述新描述子的长度是所述原始局部二值模式矢量长度的三倍。The three descriptors
Figure PCTCN2017100581-appb-000011
Descriptor norm ,
Figure PCTCN2017100581-appb-000012
A join is made to obtain a new descriptor of length 3*2^P, the length of the new descriptor being three times the length of the original local binary pattern vector.
另一方面，本发明还提供一种高光谱图像局部特征描述子的形成系统，所述系统包括：In another aspect, the present invention also provides a system for forming a local feature descriptor of a hyperspectral image, the system comprising:
特征描述模块，用于利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述；a feature description module for describing a neighborhood feature by using an angle between spectral vectors and a spectral vector;
形成描述子模块，用于根据描述结果形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。A description sub-module is formed for forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
优选的，所述特征描述模块具体用于：Preferably, the feature description module is specifically configured to:
定义
Figure PCTCN2017100581-appb-000013
为中心矢量vectorc与合成所述中心矢量vectorc的中间一维矢量的夹角值，其中，中心矢量vectorc的长度为N，中心矢量vectorc的模为normc，N为奇数，half为长度N的中间值；definition
Figure PCTCN2017100581-appb-000013
Is the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
定义
Figure PCTCN2017100581-appb-000014
为中心矢量vectorc与合成所述中心矢量vectorc的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000014
The angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
优选的，所述特征描述模块具体还用于： Preferably, the feature description module is further used to:
定义
Figure PCTCN2017100581-appb-000015
为邻域矢量vectorp与合成所述中心矢量vectorc的中间一维矢量的夹角值，其中，中心矢量vectorc的局部邻域为V，邻域半径为R，邻域点数为P，邻域矢量vectorp(p＝0,1,...,P-1)的模为normp，且邻域矢量vectorp的长度与中心矢量vectorc相同；definition
Figure PCTCN2017100581-appb-000015
Is the angle between vector values of the intermediate vector p neighborhood of the center vector of the synthesis of one-dimensional vector c is a vector, wherein the vector center vector c local neighborhood of V, neighborhood radius is R, the neighborhood points is P, o The modulus of the domain vector vector p (p=0, 1, . . . , P-1) is norm p , and the length of the neighborhood vector vector p is the same as the center vector vector c ;
定义
Figure PCTCN2017100581-appb-000016
为邻域矢量vectorp与合成所述邻域矢量vectorp的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000016
The angle between the neighborhood vector vector p and the first dimension vector of the neighborhood vector vector p is synthesized.
优选的，所述特征描述模块具体还用于：Preferably, the feature description module is further used to:
由原始LBP算子表达式得到描述邻域空间特征的第一个数值
Figure PCTCN2017100581-appb-000017
其中，
Figure PCTCN2017100581-appb-000018
The first value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000017
among them,
Figure PCTCN2017100581-appb-000018
由原始LBP算子表达式得到描述邻域空间特征的第二个数值
Figure PCTCN2017100581-appb-000019
The second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000019
由原始LBP算子表达式得到描述邻域空间特征的第三个数值
Figure PCTCN2017100581-appb-000020
The third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000020
优选的，所述形成描述子模块具体用于：Preferably, the forming description submodule is specifically configured to:
分别将中心矢量vectorc和邻域矢量vectorp(p＝0,1,...,P-1)的ARC0、norm、ARC1进行比较，以得到长度为2^P三种描述子
Figure PCTCN2017100581-appb-000021
descriptornorm、
Figure PCTCN2017100581-appb-000022
Comparing the central vector vector c with the ARC 0 , norm , and ARC 1 of the neighborhood vector vector p (p=0, 1, . . . , P-1), respectively, to obtain three descriptors of length 2^P.
Figure PCTCN2017100581-appb-000021
Descriptor norm ,
Figure PCTCN2017100581-appb-000022
将所述三种描述子
Figure PCTCN2017100581-appb-000023
descriptornorm、
Figure PCTCN2017100581-appb-000024
进行联结，以得到长度为3*2^P的新描述子，所述新描述子的长度是所述原始局部二值模式矢量长度的三倍。The three descriptors
Figure PCTCN2017100581-appb-000023
Descriptor norm ,
Figure PCTCN2017100581-appb-000024
A join is made to obtain a new descriptor of length 3*2^P, the length of the new descriptor being three times the length of the original local binary pattern vector.
本发明提供的技术方案利用光谱矢量间的夹角和光谱矢量的模对邻域特征的特征描述作用，形成一个矢量长度是原始局部二值模式矢量长度的三倍的描 述子，同时结合了高光谱特征点邻域的空间位置信息和光谱矢量信息，能适用于高光谱图像。The technical solution provided by the invention utilizes the angle between the spectral vectors and the characterization of the neighborhood features of the spectral vector to form a vector whose length is three times the length of the original local binary pattern vector. The narration, combined with the spatial position information and spectral vector information of the neighborhood of hyperspectral feature points, can be applied to hyperspectral images.
附图说明DRAWINGS
图1为本发明一实施方式中高光谱图像局部特征描述子的形成方法流程图；1 is a flow chart showing a method for forming a local feature descriptor of a hyperspectral image according to an embodiment of the present invention;
图2为本发明一实施方式中中心光谱矢量的邻域表示图；2 is a neighborhood representation of a central spectral vector in accordance with an embodiment of the present invention;
图3为本发明一实施方式中高光谱图像局部特征描述子的形成系统10的内部结构示意图。FIG. 3 is a schematic diagram showing the internal structure of a forming system 10 for a local feature descriptor of a hyperspectral image according to an embodiment of the present invention.
具体实施方式Detailed ways
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
本发明具体实施方式提供了一种高光谱图像局部特征描述子的形成方法，其中，所述方法主要包括如下步骤：A specific embodiment of the present invention provides a method for forming a local feature descriptor of a hyperspectral image, wherein the method mainly includes the following steps:
特征描述步骤：利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述；Feature description step: using the angle between the spectral vectors and the modulus of the spectral vector to describe the neighborhood features;
形成描述子步骤：根据描述结果形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。Forming a description sub-step: forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
本发明提供的一种高光谱图像局部特征描述子的形成方法，利用光谱矢量间的夹角和光谱矢量的模对邻域特征的特征描述作用，形成一个矢量长度是原始局部二值模式(Local Binary Patterns，LBP)矢量长度的三倍的描述子，同时结合了高光谱特征点邻域的空间位置信息和光谱矢量信息，能适用于高光谱图像。The invention provides a method for forming a local feature descriptor of a hyperspectral image, which uses the angle between the spectral vectors and the model of the spectral vector to describe the feature of the neighborhood feature to form a vector length which is the original local binary mode (Local Binary Patterns (LBP) triple the descriptor of the vector length, combined with the spatial position information and spectral vector information of the hyperspectral feature point neighborhood, can be applied to hyperspectral images.
以下将对本发明所提供的一种高光谱图像局部特征描述子的形成方法进行 详细说明。The method for forming a local feature descriptor of a hyperspectral image provided by the present invention will be described below. Detailed description.
请参阅图1，为本发明一实施方式中高光谱图像局部特征描述子的形成方法流程图。Please refer to FIG. 1 , which is a flowchart of a method for forming a local feature descriptor of a hyperspectral image according to an embodiment of the present invention.
在步骤S1中，特征描述步骤：利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述。In step S1, the feature description step describes the neighborhood feature by using the angle between the spectral vectors and the mode of the spectral vector.
在本实施方式中，所述特征描述步骤具体包括：In this embodiment, the feature description step specifically includes:
定义
Figure PCTCN2017100581-appb-000025
为中心矢量vectorc与合成所述中心矢量vectorc的中间一维矢量的夹角值，其中，中心矢量vectorc的长度为N，中心矢量vectorc的模为normc，N为奇数，half为长度N的中间值；definition
Figure PCTCN2017100581-appb-000025
Is the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
定义
Figure PCTCN2017100581-appb-000026
为中心矢量vectorc与合成所述中心矢量vectorc的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000026
The angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
在本实施方式中，与传统的灰度图像和RGB图像相比而言，高光谱图像包含了另一种方面的信息，即物体的光谱响应，由于高光谱图像的光谱矢量包含丰富的特征信息，所以把灰度值换成光谱矢量进行比较更具高效性，为了便于描述图2画出了中心光谱矢量的邻域表示图，其中201所指示的线条即为中心光谱矢量。In the present embodiment, compared with the conventional grayscale image and the RGB image, the hyperspectral image contains another aspect of information, that is, the spectral response of the object, since the spectral vector of the hyperspectral image contains rich feature information. Therefore, it is more efficient to replace the gray value with the spectral vector. For convenience of description, FIG. 2 draws a neighborhood representation of the central spectral vector, where the line indicated by 201 is the central spectral vector.
请参阅图2，所示为本发明一实施方式中中心光谱矢量的邻域表示图。Referring to FIG. 2, there is shown a neighborhood representation of a central spectral vector in accordance with an embodiment of the present invention.
图2中经过正方体的中心且箭头向右的表示中心光谱矢量，选取高光谱图像的光谱域中一个特征点为中心像元，这个中心像元(xc,yc,λc)的DN值和对应的前后波段(即以
Figure PCTCN2017100581-appb-000027
为间隔)上DN值组成一组矢量vc，同理这个局部邻域的邻域像元的DN值可以组成一系列矢量vp。In Fig. 2, the center of the cube and the arrow to the right represent the central spectral vector, and a feature point in the spectral domain of the hyperspectral image is selected as the central pixel, and the DN value of the central pixel (x c , y c , λ c ) And corresponding front and rear bands (ie
Figure PCTCN2017100581-appb-000027
For the interval, the upper DN value constitutes a set of vectors v c , and the DN value of the neighborhood cell of the local neighborhood can be composed of a series of vectors v p .
请继续参阅图1，在本实施方式中，所述特征描述步骤具体还包括：Continuing to refer to FIG. 1 , in the embodiment, the feature description step specifically includes:
定义
Figure PCTCN2017100581-appb-000028
为邻域矢量vectorp与合成所述领域矢量 vectorp的中间一维矢量的夹角值，其中，中心矢量vectorc的局部邻域为V，邻域半径为R，邻域点数为P，邻域矢量vectorp(p＝0,1,...,P-1)的模为normp，且邻域矢量vectorp的长度与中心矢量vectorc相同；definition
Figure PCTCN2017100581-appb-000028
The angle between the neighborhood vector vector p and the intermediate one-dimensional vector of the domain vector p , wherein the local vector vector c has a local neighborhood of V, a neighborhood radius of R, and a neighborhood point of P, neighbors. The modulus of the domain vector vector p (p=0, 1, . . . , P-1) is norm p , and the length of the neighborhood vector vector p is the same as the center vector vector c ;
定义
Figure PCTCN2017100581-appb-000029
为邻域矢量vectorp与合成所述领域矢量vectorp的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000029
A first neighborhood dimensional vector with vector vector p art of the synthesis vector of the vector p value angle.
在本实施方式中，所述特征描述步骤具体还包括：In this embodiment, the feature description step specifically includes:
由原始LBP算子表达式得到描述邻域空间特征的第一个数值
Figure PCTCN2017100581-appb-000030
其中，
Figure PCTCN2017100581-appb-000031
The first value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000030
among them,
Figure PCTCN2017100581-appb-000031
由原始LBP算子表达式得到描述邻域空间特征的第二个数值
Figure PCTCN2017100581-appb-000032
The second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000032
由原始LBP算子表达式得到描述邻域空间特征的第三个数值
Figure PCTCN2017100581-appb-000033
The third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000033
在步骤S2中，形成描述子步骤：根据描述结果形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。In step S2, a description sub-step is formed: a descriptor whose vector length is three times the length of the original local binary pattern vector is formed according to the description result.
在本实施方式中，所述形成描述子步骤具体包括：In this embodiment, the forming the sub-steps specifically includes:
分别将中心矢量vectorc和邻域矢量vectorp(p＝0,1,...,P-1)的ARC0、norm、ARC1进行比较，以得到长度为2^P三种描述子
Figure PCTCN2017100581-appb-000034
descriptornorm、
Figure PCTCN2017100581-appb-000035
Comparing the central vector vector c with the ARC 0 , norm , and ARC 1 of the neighborhood vector vector p (p=0, 1, . . . , P-1), respectively, to obtain three descriptors of length 2^P.
Figure PCTCN2017100581-appb-000034
Descriptor norm ,
Figure PCTCN2017100581-appb-000035
将所述三种描述子
Figure PCTCN2017100581-appb-000036
descriptornorm、
Figure PCTCN2017100581-appb-000037
进行联结，以得到长度为3*2^P的新描述子，所述新描述子的长度是所述原始局部二值模式矢量长度的三倍。The three descriptors
Figure PCTCN2017100581-appb-000036
Descriptor norm ,
Figure PCTCN2017100581-appb-000037
A join is made to obtain a new descriptor of length 3*2^P, the length of the new descriptor being three times the length of the original local binary pattern vector.
在本实施方式中，高光谱图像将反映物质辐射属性的光谱信息与反映物体空间几何关系的二维图像信息有机地结合在一起，使得高光谱图像能够比灰度 图像和彩色图像提供更多的信息，本发明提出的这种“图谱合一”的图像结合了二维图像和光谱信息各自的优点，拓宽了图像的分析方法，对于图像分析与识别有着十分重要的意义。In the present embodiment, the hyperspectral image organically combines the spectral information reflecting the radiation property of the substance with the two-dimensional image information reflecting the geometric relationship of the object space, so that the hyperspectral image can be grayscaled. The image and the color image provide more information. The image of the "integration of the map" proposed by the present invention combines the advantages of the two-dimensional image and the spectral information, and broadens the analysis method of the image, which is very important for image analysis and recognition. The meaning.
本发明提供的一种高光谱图像局部特征描述子的形成方法，针对高光谱图像的光谱矢量具有特征属性这一特点，提出了一种全新的基于局部二值模式的描述子，在这种描述子中利用光谱矢量间的夹角和光谱矢量的模对邻域特征的特征描述作用，形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。该算法中考虑到高光谱图像的特征，同时结合了高光谱特征点邻域的空间位置信息和光谱矢量信息。The invention provides a method for forming a local feature descriptor of a hyperspectral image, and proposes a novel descriptor based on a local binary pattern for the feature of the spectral vector of the hyperspectral image. In the sub-use, the angle between the spectral vectors and the model of the spectral vector are used to describe the features of the neighborhood, and a descriptor whose vector length is three times the length of the original local binary pattern vector is formed. The algorithm takes into account the characteristics of the hyperspectral image and combines the spatial position information and spectral vector information of the neighborhood of the hyperspectral feature points.
本发明具体实施方式还提供一种高光谱图像局部特征描述子的形成系统10，主要包括：A specific embodiment of the present invention further provides a system for forming a local feature descriptor of a hyperspectral image, which mainly includes:
特征描述模块，用于利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述；a feature description module for describing a neighborhood feature by using an angle between spectral vectors and a spectral vector;
形成描述子模块，用于根据描述结果形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。A description sub-module is formed for forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
本发明提供的一种高光谱图像局部特征描述子的形成系统10，利用光谱矢量间的夹角和光谱矢量的模对邻域特征的特征描述作用，形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子，同时结合了高光谱特征点邻域的空间位置信息和光谱矢量信息，能适用于高光谱图像。The present invention provides a hyperspectral image local feature descriptor forming system 10, which utilizes the angle between the spectral vectors and the spectral vector to model the feature description of the neighborhood feature to form a vector length which is the original local binary pattern vector. The descriptor of three times the length, combined with the spatial position information and spectral vector information of the neighborhood of the hyperspectral feature points, can be applied to hyperspectral images.
请参阅图3，所示为本发明一实施方式中高光谱图像局部特征描述子的形成系统10的结构示意图。Referring to FIG. 3, a schematic structural diagram of a system 10 for forming a local feature descriptor of a hyperspectral image according to an embodiment of the present invention is shown.
在本实施方式中，高光谱图像局部特征描述子的形成系统10，主要包括特征描述模块11和形成描述子模块12。In the present embodiment, the hyperspectral image local feature descriptor forming system 10 mainly includes a feature description module 11 and a formation description sub-module 12.
特征描述模块11，用于利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述。The feature description module 11 is configured to describe the neighborhood features by using the angle between the spectral vectors and the modulus of the spectral vector.
在本实施方式中，所述特征描述模块11具体用于： In this embodiment, the feature description module 11 is specifically configured to:
定义
Figure PCTCN2017100581-appb-000038
为中心矢量vectorc与合成所述中心矢量vectorc的中间一维矢量的夹角值，其中，中心矢量vectorc的长度为N，中心矢量vectorc的模为normc，N为奇数，half为长度N的中间值；definition
Figure PCTCN2017100581-appb-000038
Is the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
定义
Figure PCTCN2017100581-appb-000039
为中心矢量vectorc与合成所述中心矢量vectorc的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000039
The angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
在本实施方式中，所述特征描述模块11具体还用于：In this embodiment, the feature description module 11 is further specifically configured to:
定义
Figure PCTCN2017100581-appb-000040
为邻域矢量vectorp与合成所述领域矢量vectorp的中间一维矢量的夹角值，其中，中心矢量vectorc的局部邻域为V，邻域半径为R，邻域点数为P，邻域矢量vectorp(p＝0,1,...,P-1)的模为normp，且邻域矢量vectorp的长度与中心矢量vectorc相同；definition
Figure PCTCN2017100581-appb-000040
The angle between the neighborhood vector vector p and the intermediate one-dimensional vector of the domain vector p , wherein the local vector vector c has a local neighborhood of V, a neighborhood radius of R, and a neighborhood point of P, neighbors. The modulus of the domain vector vector p (p=0, 1, . . . , P-1) is norm p , and the length of the neighborhood vector vector p is the same as the center vector vector c ;
定义
Figure PCTCN2017100581-appb-000041
为邻域矢量vectorp与合成所述领域矢量vectorp的第一维矢量的夹角值。definition
Figure PCTCN2017100581-appb-000041
A first neighborhood dimensional vector with vector vector p art of the synthesis vector of the vector p value angle.
在本实施方式中，所述特征描述模块11具体还用于：In this embodiment, the feature description module 11 is further specifically configured to:
由原始LBP算子表达式得到描述邻域空间特征的第一个数值
Figure PCTCN2017100581-appb-000042
其中，
Figure PCTCN2017100581-appb-000043
The first value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000042
among them,
Figure PCTCN2017100581-appb-000043
由原始LBP算子表达式得到描述邻域空间特征的第二个数值
Figure PCTCN2017100581-appb-000044
The second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000044
由原始LBP算子表达式得到描述邻域空间特征的第三个数值
Figure PCTCN2017100581-appb-000045
The third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
Figure PCTCN2017100581-appb-000045
形成描述子模块12，用于根据描述结果形成一个矢量长度是原始局部二值 模式矢量长度的三倍的描述子。Forming a description sub-module 12 for forming a vector length according to the description result is an original local binary value A three-dimensional descriptor of the length of the pattern vector.
在本实施方式中，所述形成描述子模块12具体用于：In this embodiment, the forming description sub-module 12 is specifically configured to:
分别将中心矢量vectorc和邻域矢量vectorp(p＝0,1,...,P-1)的ARC0、norm、ARC1进行比较，以得到长度为2^P三种描述子
Figure PCTCN2017100581-appb-000046
descriptornorm、
Figure PCTCN2017100581-appb-000047
Comparing the central vector vector c with the ARC 0 , norm , and ARC 1 of the neighborhood vector vector p (p=0, 1, . . . , P-1), respectively, to obtain three descriptors of length 2^P.
Figure PCTCN2017100581-appb-000046
Descriptor norm ,
Figure PCTCN2017100581-appb-000047
将所述三种描述子
Figure PCTCN2017100581-appb-000048
descriptornorm、
Figure PCTCN2017100581-appb-000049
进行联结，以得到长度为3*2^P的新描述子，所述新描述子的长度是所述原始局部二值模式矢量长度的三倍。The three descriptors
Figure PCTCN2017100581-appb-000048
Descriptor norm ,
Figure PCTCN2017100581-appb-000049
A join is made to obtain a new descriptor of length 3*2^P, the length of the new descriptor being three times the length of the original local binary pattern vector.
本发明提供的一种高光谱图像局部特征描述子的形成系统10，利用光谱矢量间的夹角和光谱矢量的模对邻域特征的特征描述作用，形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子，同时结合了高光谱特征点邻域的空间位置信息和光谱矢量信息，能适用于高光谱图像。The present invention provides a hyperspectral image local feature descriptor forming system 10, which utilizes the angle between the spectral vectors and the spectral vector to model the feature description of the neighborhood feature to form a vector length which is the original local binary pattern vector. The descriptor of three times the length, combined with the spatial position information and spectral vector information of the neighborhood of the hyperspectral feature points, can be applied to hyperspectral images.
值得注意的是，上述实施例中，所包括的各个单元只是按照功能逻辑进行划分的，但并不局限于上述的划分，只要能够实现相应的功能即可；另外，各功能单元的具体名称也只是为了便于相互区分，并不用于限制本发明的保护范围。It should be noted that, in the foregoing embodiment, each unit included is only divided according to functional logic, but is not limited to the above division, as long as the corresponding function can be implemented; in addition, the specific name of each functional unit is also They are only used to facilitate mutual differentiation and are not intended to limit the scope of the present invention.
另外，本领域普通技术人员可以理解实现上述各实施例方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，相应的程序可以存储于一计算机可读取存储介质中，所述的存储介质，如ROM/RAM、磁盘或光盘等。In addition, those skilled in the art can understand that all or part of the steps of implementing the above embodiments may be completed by a program to instruct related hardware, and the corresponding program may be stored in a computer readable storage medium. Storage medium, such as ROM/RAM, disk or CD.
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。 The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection of the present invention. Within the scope.

Claims (10)

  1. 一种高光谱图像局部特征描述子的形成方法，其特征在于，所述方法包括：A method for forming a local feature descriptor of a hyperspectral image, characterized in that the method comprises:
    特征描述步骤：利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述；Feature description step: using the angle between the spectral vectors and the modulus of the spectral vector to describe the neighborhood features;
    形成描述子步骤：根据描述结果形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。Forming a description sub-step: forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
  2. 如权利要求1所述的高光谱图像局部特征描述子的形成方法，其特征在于，所述特征描述步骤具体包括：The method for forming a local feature descriptor of a hyperspectral image according to claim 1, wherein the step of describing the feature specifically comprises:
    定义
    Figure PCTCN2017100581-appb-100001
    为中心矢量vectorc与合成所述中心矢量vectorc的中间一维矢量的夹角值，其中，中心矢量vectorc的长度为N，中心矢量vectorc的模为normc，N为奇数，half为长度N的中间值；    definition
    Figure PCTCN2017100581-appb-100001
    Is the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
    定义
    Figure PCTCN2017100581-appb-100002
    为中心矢量vectorc与合成所述中心矢量vectorc的第一维矢量的夹角值。    definition
    Figure PCTCN2017100581-appb-100002
    The angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
  3. 如权利要求2所述的高光谱图像局部特征描述子的形成方法，其特征在于，所述特征描述步骤具体还包括：The method for forming a local feature descriptor of a hyperspectral image according to claim 2, wherein the feature description step further comprises:
    定义
    Figure PCTCN2017100581-appb-100003
    为邻域矢量vectorp与合成所述邻域矢量vectorp的中间一维矢量的夹角值，其中，中心矢量vectorc的局部邻域为V，邻域半径为R，邻域点数为P，邻域矢量vectorp(p＝0,1,...,P-1)的模为normp，且邻域矢量vectorp的长度N′与中心矢量vectorc的长度N相同，half′为长度N′中间值；    definition
    Figure PCTCN2017100581-appb-100003
    Vector for the neighborhood in the synthesis of the vector p o value of the intermediate field vector angle of a vector p-dimensional vector, wherein the vector center vector c local neighborhood of V, neighborhood radius is R, the neighborhood points is P, The modulus of the neighborhood vector vector p (p=0,1,...,P-1) is norm p , and the length N′ of the neighborhood vector vector p is the same as the length N of the center vector vector c , and the half′ is the length. N' intermediate value;
    定义
    Figure PCTCN2017100581-appb-100004
    为邻域矢量vectorp与合成所述邻域矢量vectorp的第一维矢量的夹角值。    definition
    Figure PCTCN2017100581-appb-100004
    The angle between the neighborhood vector vector p and the first dimension vector of the neighborhood vector vector p is synthesized.
  4. 如权利要求3所述的高光谱图像局部特征描述子的形成方法，其特征在 于，所述特征描述步骤具体还包括：A method for forming a local feature descriptor of a hyperspectral image according to claim 3, characterized in that The feature description step specifically includes:
    由原始LBP算子表达式得到描述邻域空间特征的第一个数值
    Figure PCTCN2017100581-appb-100005
    其中，
    Figure PCTCN2017100581-appb-100006
    The first value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
    Figure PCTCN2017100581-appb-100005
    among them,
    Figure PCTCN2017100581-appb-100006
    由原始LBP算子表达式得到描述邻域空间特征的第二个数值
    Figure PCTCN2017100581-appb-100007
    The second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
    Figure PCTCN2017100581-appb-100007
    由原始LBP算子表达式得到描述邻域空间特征的第三个数值
    Figure PCTCN2017100581-appb-100008
    The third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
    Figure PCTCN2017100581-appb-100008
  5. 如权利要求4所述的高光谱图像局部特征描述子的形成方法，其特征在于，所述形成描述子步骤具体包括：The method for forming a local feature descriptor of a hyperspectral image according to claim 4, wherein the forming the sub-step comprises:
    分别将中心矢量vectorc和邻域矢量vectorp(p＝0,1,...,P-1)的ARC0、norm、ARC1进行比较，以得到长度为2^P三种描述子
    Figure PCTCN2017100581-appb-100009
    descriptornorm、
    Figure PCTCN2017100581-appb-100010
    Comparing the central vector vector c with the ARC 0 , norm , and ARC 1 of the neighborhood vector vector p (p=0, 1, . . . , P-1), respectively, to obtain three descriptors of length 2^P.
    Figure PCTCN2017100581-appb-100009
    Descriptor norm ,
    Figure PCTCN2017100581-appb-100010
    将所述三种描述子
    Figure PCTCN2017100581-appb-100011
    descriptornorm、
    Figure PCTCN2017100581-appb-100012
    进行联结，以得到长度为3*2^P的新描述子，所述新描述子的长度是所述原始局部二值模式矢量长度的三倍。    The three descriptors
    Figure PCTCN2017100581-appb-100011
    Descriptor norm ,
    Figure PCTCN2017100581-appb-100012
    A join is made to obtain a new descriptor of length 3*2^P, the length of the new descriptor being three times the length of the original local binary pattern vector.
  6. 一种高光谱图像局部特征描述子的形成系统，其特征在于，所述系统包括：A system for forming a local feature descriptor of a hyperspectral image, characterized in that the system comprises:
    特征描述模块，用于利用光谱矢量间的夹角和光谱矢量的模对邻域特征进行描述；a feature description module for describing a neighborhood feature by using an angle between spectral vectors and a spectral vector;
    形成描述子模块，用于根据描述结果形成一个矢量长度是原始局部二值模式矢量长度的三倍的描述子。A description sub-module is formed for forming a descriptor whose vector length is three times the length of the original local binary pattern vector according to the description result.
  7. 如权利要求6所述的高光谱图像局部特征描述子的形成系统，其特征在于，所述特征描述模块具体用于： The system for forming a hyperspectral image local feature descriptor according to claim 6, wherein the feature description module is specifically configured to:
    定义
    Figure PCTCN2017100581-appb-100013
    为中心矢量vectorc与合成所述中心矢量vectorc的中间一维矢量的夹角值，其中，中心矢量vectorc的长度为N，中心矢量vectorc的模为normc，N为奇数，half为长度N的中间值；    definition
    Figure PCTCN2017100581-appb-100013
    Is the angle between the center value of the intermediate vector vector c and the synthesized vector c is a vector of the center-dimensional vector, wherein the vector length of the central vector c is N, the center of the die is a vector c vector norm c, N is an odd number, half of The median of the length N;
    定义
    Figure PCTCN2017100581-appb-100014
    为中心矢量vectorc与合成所述中心矢量vectorc的第一维矢量的夹角值。    definition
    Figure PCTCN2017100581-appb-100014
    The angle between the center vector vector c and the first dimension vector of the center vector vector c is synthesized.
  8. 如权利要求7所述的高光谱图像局部特征描述子的形成系统，其特征在于，所述特征描述模块具体还用于：The method for forming a hyperspectral image local feature descriptor according to claim 7, wherein the feature description module is further configured to:
    定义
    Figure PCTCN2017100581-appb-100015
    为邻域矢量vectorp与合成所述邻域矢量vectorp的中间一维矢量的夹角值，其中，中心矢量vectorc的局部邻域为V，邻域半径为R，邻域点数为P，邻域矢量vectorp(p＝0,1,...,P-1)的模为normp，且邻域矢量vectorp的长度与中心矢量vectorc相同；    definition
    Figure PCTCN2017100581-appb-100015
    Vector for the neighborhood in the synthesis of the vector p o value of the intermediate field vector angle of a vector p-dimensional vector, wherein the vector center vector c local neighborhood of V, neighborhood radius is R, the neighborhood points is P, The modulus of the neighborhood vector vector p (p=0, 1, . . . , P-1) is norm p , and the length of the neighborhood vector vector p is the same as the center vector vector c ;
    定义
    Figure PCTCN2017100581-appb-100016
    为邻域矢量vectorp与合成所述邻域矢量vectorp的第一维矢量的夹角值。    definition
    Figure PCTCN2017100581-appb-100016
    The angle between the neighborhood vector vector p and the first dimension vector of the neighborhood vector vector p is synthesized.
  9. 如权利要求8所述的高光谱图像局部特征描述子的形成系统，其特征在于，所述特征描述模块具体还用于：The system for forming a hyperspectral image local feature descriptor according to claim 8, wherein the feature description module is further configured to:
    由原始LBP算子表达式得到描述邻域空间特征的第一个数值
    Figure PCTCN2017100581-appb-100017
    其中，
    Figure PCTCN2017100581-appb-100018
    The first value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
    Figure PCTCN2017100581-appb-100017
    among them,
    Figure PCTCN2017100581-appb-100018
    由原始LBP算子表达式得到描述邻域空间特征的第二个数值
    Figure PCTCN2017100581-appb-100019
    The second value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
    Figure PCTCN2017100581-appb-100019
    由原始LBP算子表达式得到描述邻域空间特征的第三个数值
    Figure PCTCN2017100581-appb-100020
    The third value describing the neighborhood spatial feature is obtained from the original LBP operator expression.
    Figure PCTCN2017100581-appb-100020
  10. 如权利要求9所述的高光谱图像局部特征描述子的形成系统，其特征在于，所述形成描述子模块具体用于：The system for forming a local feature descriptor of a hyperspectral image according to claim 9, wherein the forming description sub-module is specifically configured to:
    分别将中心矢量vectorc和邻域矢量vectorp(p＝0,1,...,P-1)的ARC0、norm、ARC1进行比较，以得到长度为2^P三种描述子
    Figure PCTCN2017100581-appb-100021
    descriptornorm、
    Figure PCTCN2017100581-appb-100022
    Comparing the central vector vector c with the ARC 0 , norm , and ARC 1 of the neighborhood vector vector p (p=0, 1, . . . , P-1), respectively, to obtain three descriptors of length 2^P.
    Figure PCTCN2017100581-appb-100021
    Descriptor norm ,
    Figure PCTCN2017100581-appb-100022
    将所述三种描述子
    Figure PCTCN2017100581-appb-100023
    descriptornorm、
    Figure PCTCN2017100581-appb-100024
    进行联结，以得到长度为3*2^P的新描述子，所述新描述子的长度是所述原始局部二值模式矢量长度的三倍。     The three descriptors
    Figure PCTCN2017100581-appb-100023
    Descriptor norm ,
    Figure PCTCN2017100581-appb-100024
    A join is made to obtain a new descriptor of length 3*2^P, the length of the new descriptor being three times the length of the original local binary pattern vector.
PCT/CN2017/100581 2017-09-05 2017-09-05 Method for forming local feature descriptor of hyperspectral image and forming system WO2019047025A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/100581 WO2019047025A1 (en) 2017-09-05 2017-09-05 Method for forming local feature descriptor of hyperspectral image and forming system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/100581 WO2019047025A1 (en) 2017-09-05 2017-09-05 Method for forming local feature descriptor of hyperspectral image and forming system

Publications (1)

Publication Number Publication Date
WO2019047025A1 true WO2019047025A1 (en) 2019-03-14

Family

ID=65633452

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/100581 WO2019047025A1 (en) 2017-09-05 2017-09-05 Method for forming local feature descriptor of hyperspectral image and forming system

Country Status (1)

Country Link
WO (1) WO2019047025A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110311142A1 (en) * 2010-06-18 2011-12-22 National Ict Australia Limited Descriptor of a hyperspectral or multispectral image
CN105608433A (en) * 2015-12-23 2016-05-25 北京化工大学 Nuclear coordinated expression-based hyperspectral image classification method
CN106485238A (en) * 2016-11-01 2017-03-08 深圳大学 A kind of high-spectrum remote sensing feature extraction and sorting technique and its system
CN106529508A (en) * 2016-12-07 2017-03-22 西安电子科技大学 Local and non-local multi-feature semantics-based hyperspectral image classification method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110311142A1 (en) * 2010-06-18 2011-12-22 National Ict Australia Limited Descriptor of a hyperspectral or multispectral image
CN105608433A (en) * 2015-12-23 2016-05-25 北京化工大学 Nuclear coordinated expression-based hyperspectral image classification method
CN106485238A (en) * 2016-11-01 2017-03-08 深圳大学 A kind of high-spectrum remote sensing feature extraction and sorting technique and its system
CN106529508A (en) * 2016-12-07 2017-03-22 西安电子科技大学 Local and non-local multi-feature semantics-based hyperspectral image classification method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LI, WEI.: "Local Binary Patterns and Extreme Learning Machine for Hyperspectral Imagery Classification", IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, vol. 53, no. 7, 31 July 2015 (2015-07-31), XP011575081 *
YE, ZHEN ET AL.: "Hyperspectral Image Classification Based on Principal Component Analysis and Local Binary Patterns", LASER & OPTOELECTRONICS, no. 11, 6 July 2017 (2017-07-06), pages 111006-1 - 111006-9 *

Similar Documents

Publication Publication Date Title
CN110443842B (en) Depth map prediction method based on visual angle fusion
Xie et al. Joint super resolution and denoising from a single depth image
Wang et al. Neuris: Neural reconstruction of indoor scenes using normal priors
Faridul et al. Colour mapping: A review of recent methods, extensions and applications
US7817161B2 (en) Texture synthesis using dimensionality-reduced appearance space
CN108876723B (en) Method for constructing color background of gray target image
Xie et al. Single depth image super resolution and denoising via coupled dictionary learning with local constraints and shock filtering
Pesaresi et al. A new compact representation of morphological profiles: Report on first massive VHR image processing at the JRC
US11887341B2 (en) Systems and methods for hue based encoding of a digital image
Li et al. SiamBAG: Band attention grouping-based Siamese object tracking network for hyperspectral videos
Chaki et al. Image color feature extraction techniques: fundamentals and applications
Yaman et al. An iterative adaptive multi-modal stereo-vision method using mutual information
US8520980B2 (en) Image searching method and apparatus
IlaVennila Optimization fusion approach for image segmentation using k-means algorithm
ALEjaily et al. Fusion of remote sensing images using contourlet transform
Li et al. I-gans for infrared image generation
WO2019047025A1 (en) Method for forming local feature descriptor of hyperspectral image and forming system
Obulesu et al. Image retrieval based local motif patterns code
Wang et al. Multi-sensor Image Fusion Algorithm Based on Multiresolution Analysis.
Dinesh et al. Point cloud video super-resolution via partial point coupling and graph smoothness
Amirkolaee et al. Convolutional neural network architecture for digital surface model estimation from single remote sensing image
CN107679538B (en) Method and system for forming local feature descriptor of hyperspectral image
Guerrero et al. DepthCut: improved depth edge estimation using multiple unreliable channels
Dixit et al. Copy-move image forgery detection using frequency-based techniques: a review
Silverman et al. Segmentation of multi-dimensional infrared imagery from histograms

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17924465

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 30/09/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17924465

Country of ref document: EP

Kind code of ref document: A1