CN110738215A - hyperspectral remote sensing image feature extraction device and method - Google Patents
hyperspectral remote sensing image feature extraction device and method Download PDFInfo
- Publication number
- CN110738215A CN110738215A CN201910806619.4A CN201910806619A CN110738215A CN 110738215 A CN110738215 A CN 110738215A CN 201910806619 A CN201910806619 A CN 201910806619A CN 110738215 A CN110738215 A CN 110738215A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- unit
- images
- remote sensing
- 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
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- 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
- G06V10/58—Extraction of image or video features relating to hyperspectral data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
Abstract
The invention discloses an hyperspectral remote sensing image feature extraction device and a method, which comprises an image acquisition unit for acquiring images, an image classification unit for classifying the images acquired by the image acquisition unit, a data analysis unit for analyzing the images of the image classification unit, wherein the data analysis unit comprises a principal component analysis module, a pixel purity analysis module and a minimum noise separation module, the images are acquired by the image acquisition unit, the acquired images are sent to the image classification unit by the image acquisition unit, the images are classified by the image classification unit, the principal component and pixel purity of the images are analyzed and the minimum noise is separated by the data analysis unit, the images are classified by the image classification unit, the image data which are not pure pixels are subjected to pixel unmixing and then are classified and identified, and meanwhile, the data analysis unit can extract features such as principal component, pixel purity, minimum noise separation change and the like.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an hyperspectral remote sensing image feature extraction device and method.
Background
A large number of wave bands of the hyperspectral remote sensing images provide abundant spectrum, radiation and geometric information for ground feature information extraction, and are beneficial to more precise ground feature classification and target identification.
The existing method for extracting the hyperspectral features mainly comprises the steps of extracting the spatial features, wherein the spatial feature extraction mainly utilizes spatial information of different wave bands to express the hyperspectral remote sensing images, firstly extracts the spatial features of the wave bands, then superposes the spatial features of the different wave bands on , of the existing hyperspectral feature extraction generally needs to adopt pure pixel assumptions to analyze, and pure pixel assumptions are adopted for data with large noise or high mixing to hardly obtain end member extraction results meeting requirements.
Disclosure of Invention
The invention aims to overcome the problems in the prior art, and provides hyperspectral remote sensing image feature extraction devices and methods.
In order to achieve the technical purpose and achieve the technical effect, the invention is realized by the following technical scheme:
A hyperspectral remote sensing image feature extraction device comprises:
the image acquisition unit is used for acquiring images;
the image classification unit is used for classifying the images acquired by the image acquisition unit;
and the data analysis unit is used for image analysis of the image classification unit.
Further , the data analysis unit includes a principal component analysis module, a pixel purity analysis module, and a minimum noise separation module.
method for extracting the characteristics of the hyperspectral remote sensing image comprises the following steps.
Collecting the image through an image collecting unit;
the image acquisition unit sends the acquired images to the image classification unit, and the image classification unit classifies the images;
and the data analysis unit analyzes the principal component and the pixel purity of the image and separates the minimum noise.
, sending the image data collected by the image collecting unit to an image classifying unit, preprocessing the image by the image classifying unit, detecting whether the image data is pure image elements, if so, extracting the characteristics, then classifying and identifying, if not, extracting the characteristics, then unmixing the image elements, and finally classifying and identifying.
, selecting n classified data to perform principal component analysis, which follows the following formula
Fn=a11X1+a21X1+……+an1Xn
………
Fn=a1nX1+a2nX2+……+annXn。
, selecting the hyperspectral image data of n classified pixels, and representing the n data as X = [ X ]1,x2,…,xn],xi(i =1, 2, …, n) is a D-dimensional vector, and each pixel spectrum is divided into m endsElements (j =1, 2, …, m) are linearly mixed, and the pixel spectra and the end-member spectra are substituted into the following formula
The ratio of end members in the pixel is shown in the formula, and the deviation of the ith pixel is shown.
Further , assuming that the vector formed by the ith wave band in the hyperspectral image is composed of the sum of the noiseless vector signals under ideal conditions, it can be represented by the following formula
The noisy image n can be separated from the original image z and then the covariance matrix Q of z and n is solvedzAnd QnAnd finally, extracting the characteristics of the minimum noise separation change.
The benefit effects of the invention are:
the image classification unit is used for classifying the images, the image classification unit can be used for preprocessing the images, pixel unmixing is carried out on image data which are not pure pixels, classification and identification are carried out, and meanwhile, the data analysis unit can be used for extracting characteristics such as principal components, pixel purity and minimum noise separation change.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a schematic diagram of an image classification process according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only partial embodiments of of the present invention, rather than all embodiments.
As shown in fig. 1-2, the invention provides an kinds of hyperspectral remote sensing image feature extraction device, which comprises:
the image acquisition unit is used for acquiring images;
the image classification unit is used for classifying the images acquired by the image acquisition unit;
and the data analysis unit is used for image analysis of the image classification unit.
The data analysis unit comprises a principal component analysis module, a pixel purity analysis module and a minimum noise separation module.
method for extracting the characteristics of the hyperspectral remote sensing image comprises the following steps.
Collecting the image through an image collecting unit;
the image acquisition unit sends the acquired images to the image classification unit, and the image classification unit classifies the images;
and the data analysis unit analyzes the principal component and the pixel purity of the image and separates the minimum noise.
The image classification unit is used for preprocessing an image, detecting whether the image data is a pure pixel or not, extracting and then classifying and identifying the feature if the image data is the pure pixel, extracting and then de-mixing the pixels if the image data is not the pure pixel, and finally classifying and identifying the image.
Wherein, n classified data are selected for principal component analysis, and the principal component analysis follows the following formula
Fn=a11X1+a21X1+……+an1Xn
………
Fn=a1nX1+a2nX2+……+annXn。
, selecting the hyperspectral image data of n classified pixels, and representing the n data as X = [ X ]1,x2,…,xn],xi(i =1, 2, …, n) is a D-dimensional vector, each pixel spectra is formed by linearly mixing m end members (j =1, 2, …, m), and the pixel spectra and the end member spectra are brought into the following formula
The ratio of end members in the pixel is shown in the formula, and the deviation of the ith pixel is shown.
Wherein, if the vector formed by the ith wave band in the hyperspectral image is composed of the sum of noiseless vector signals under an ideal condition, the vector can be expressed by the following formula
The noisy image n can be separated from the original image z and then the covariance matrix Q of z and n is solvedzAnd QnAnd finally, extracting the characteristics of the minimum noise separation change.
In summary, the image classification unit classifies the images, the image classification unit can preprocess the images, pixel unmixing is carried out on the image data which are not pure pixels, then classification and identification are carried out, and meanwhile, the data analysis unit can extract the characteristics of principal components, pixel purity, minimum noise separation change and the like.
In the description herein, reference to the term " embodiments," "examples," "specific examples," or the like, means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least embodiments or examples of the invention.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (7)
1, kinds of high spectrum remote sensing image feature extraction element, its characterized in that includes:
the image acquisition unit is used for acquiring images;
the image classification unit is used for classifying the images acquired by the image acquisition unit;
and the data analysis unit is used for image analysis of the image classification unit.
2. The kinds of high spectrum remote sensing image feature extraction device of claim 1, wherein the data analysis unit comprises a principal component analysis module, a pixel purity analysis module and a minimum noise separation module.
3, A hyperspectral remote sensing image feature extraction method, which is characterized by comprising:
collecting the image through an image collecting unit;
the image acquisition unit sends the acquired images to the image classification unit, and the image classification unit classifies the images;
and the data analysis unit analyzes the principal component and the pixel purity of the image and separates the minimum noise.
4. The method for extracting the characteristics of the hyperspectral remote sensing images according to claim 3 is characterized in that the image data collected by the image collection unit is sent to an image classification unit, the image classification unit preprocesses the image and detects whether the image data is a pure pixel or not, if the image data is a pure pixel, the characteristics are extracted and then classified and identified, and if the image data is not a pure pixel, the characteristics are extracted and then pixel unmixing is performed, and finally classified and identified.
5. The method for extracting the features of the hyperspectral remote sensing images according to claim 3, wherein n classified data are selected for principal component analysis, and the principal component analysis follows the following formula
Fn=a11X1+a21X1+……+an1Xn
………
Fn=a1nX1+a2nX2+……+annXn。
6. The method for extracting the characteristics of the hyperspectral remote sensing images according to claim 3, wherein the hyperspectral image data of n classified pixels are selected, and the n data are expressed as X = [ X ]1,x2,…,xn],xi(i =1, 2, …, n) is a D-dimensional vector, each pixel spectra is formed by linearly mixing m end members (j =1, 2, …, m), and the pixel spectra and the end member spectra are brought into the following formula
The ratio of end members in the pixel is shown in the formula, and the deviation of the ith pixel is shown.
7. The method for extracting features of hyperspectral remote sensing images according to claim 3, wherein if the vector formed by the ith wave band in the hyperspectral image is composed of a sum of noiseless vector signals under ideal conditions, the method can be expressed by the following formula
The noisy image n can be separated from the original image z and then the covariance matrix Q of z and n is solvedzAnd QnAnd finally, extracting the characteristics of the minimum noise separation change.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910806619.4A CN110738215A (en) | 2019-08-29 | 2019-08-29 | hyperspectral remote sensing image feature extraction device and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910806619.4A CN110738215A (en) | 2019-08-29 | 2019-08-29 | hyperspectral remote sensing image feature extraction device and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110738215A true CN110738215A (en) | 2020-01-31 |
Family
ID=69267411
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910806619.4A Pending CN110738215A (en) | 2019-08-29 | 2019-08-29 | hyperspectral remote sensing image feature extraction device and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110738215A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116026787A (en) * | 2023-03-29 | 2023-04-28 | 湖南汇湘轩生物科技股份有限公司 | Essence grade detection method and system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101109658A (en) * | 2007-06-25 | 2008-01-23 | 北京航空航天大学 | High spectrum mineral map plotting method |
CN101853506A (en) * | 2010-05-27 | 2010-10-06 | 西北工业大学 | High optical spectrum image end member extraction method based on optimized search strategy |
CN104751181A (en) * | 2015-04-02 | 2015-07-01 | 山东大学 | High spectral image Deming method based on relative abundance |
CA2939304A1 (en) * | 2016-08-17 | 2018-02-17 | Zhe Jun Wang | A pixel-based universal image info extraction models and process flow |
-
2019
- 2019-08-29 CN CN201910806619.4A patent/CN110738215A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101109658A (en) * | 2007-06-25 | 2008-01-23 | 北京航空航天大学 | High spectrum mineral map plotting method |
CN101853506A (en) * | 2010-05-27 | 2010-10-06 | 西北工业大学 | High optical spectrum image end member extraction method based on optimized search strategy |
CN104751181A (en) * | 2015-04-02 | 2015-07-01 | 山东大学 | High spectral image Deming method based on relative abundance |
CA2939304A1 (en) * | 2016-08-17 | 2018-02-17 | Zhe Jun Wang | A pixel-based universal image info extraction models and process flow |
Non-Patent Citations (1)
Title |
---|
杨可明;陈云浩;郭达志;蒋金豹;: "基于端元提取的高光谱影像特定目标识别" * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116026787A (en) * | 2023-03-29 | 2023-04-28 | 湖南汇湘轩生物科技股份有限公司 | Essence grade detection method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019169816A1 (en) | Deep neural network for fine recognition of vehicle attributes, and training method thereof | |
CN102214298B (en) | Method for detecting and identifying airport target by using remote sensing image based on selective visual attention mechanism | |
CN107239759B (en) | High-spatial-resolution remote sensing image transfer learning method based on depth features | |
CN110348357B (en) | Rapid target detection method based on deep convolutional neural network | |
CN107092884B (en) | Rapid coarse-fine cascade pedestrian detection method | |
CN111797653A (en) | Image annotation method and device based on high-dimensional image | |
CN102496157A (en) | Image detection method based on Gaussian multi-scale transform and color complexity | |
CN107480679B (en) | Road network extraction method based on classification and connected region analysis of convolutional neural network | |
CN110659649A (en) | Image processing and character recognition algorithm based on near infrared light imaging | |
CN109190456B (en) | Multi-feature fusion overlook pedestrian detection method based on aggregated channel features and gray level co-occurrence matrix | |
CN103530648A (en) | Face recognition method based on multi-frame images | |
CN103440035A (en) | Gesture recognition system in three-dimensional space and recognition method thereof | |
Masood et al. | Plants disease segmentation using image processing | |
CN112257711B (en) | Method for detecting damage fault of railway wagon floor | |
Syrjasuo et al. | Numeric image features for detection of aurora | |
CN111259756A (en) | Pedestrian re-identification method based on local high-frequency features and mixed metric learning | |
CN109800762A (en) | A kind of fuzzy license plate recognizer based on the Dynamic Matching factor | |
CN110738215A (en) | hyperspectral remote sensing image feature extraction device and method | |
Qin et al. | Video scene text frames categorization for text detection and recognition | |
CN102129569A (en) | Equipment and method for detecting object based on multiscale comparison characteristic | |
CN104463182A (en) | NBI gastroscope image processing method based on key point detection | |
CN112699901A (en) | Plant species identification system based on Internet of things | |
CN116738271A (en) | Target autonomous identification and classification method and system based on radar and video fusion | |
Chumuang et al. | Sorting Red and Green Chilies by Digital Image Processing | |
CN107341456B (en) | Weather sunny and cloudy classification method based on single outdoor color image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |