CN101140325A - Method for enhancing distinguishability cooperated with space-optical spectrum information of high optical spectrum image - Google Patents
Method for enhancing distinguishability cooperated with space-optical spectrum information of high optical spectrum image Download PDFInfo
- Publication number
- CN101140325A CN101140325A CNA2007101444495A CN200710144449A CN101140325A CN 101140325 A CN101140325 A CN 101140325A CN A2007101444495 A CNA2007101444495 A CN A2007101444495A CN 200710144449 A CN200710144449 A CN 200710144449A CN 101140325 A CN101140325 A CN 101140325A
- Authority
- CN
- China
- Prior art keywords
- information
- spatial
- spectral
- resolution
- image
- 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
Links
Images
Landscapes
- Image Processing (AREA)
Abstract
A method to jointly improve resolution of high spectral image space and spectral information relates to a method to improve spatial resolution through high spectral image information, which removes a failure to make full use of spatial information and spectral information to improve image resolution during high spectral image processing and comprises steps below: I. Inputting high spectral image data; A. Withdrawing spatial information; A I. Selecting characteristic wave band; A II. Analyzing and judging partial space; B. Withdrawing spectral information; B I. Withdrawing spectral terminal element; B II. Mixing pixel decomposition; II. Fulfilling collaborative super resolution of space and spectrum; III. Obtaining high spectral images with improved resolution. The present invention realizes breakthrough of spatial resolution during image acquisition, utilizes mixed partial relevance supporting vector mechanical decomposition to conduct spatial and spectral information collaboration technology, improve spatial resolution of high spectral image, greatly increase target detecting and locating capacity, break through limits of image acquisition means and make up hardware defects.
Description
Technical Field
The invention relates to a method for improving spatial resolution by utilizing hyperspectral image information.
Background
The spatial resolution is simply a measure of the image detail resolution capability of the imaging system, is also an index of the target subtlety in the image, represents the scene information detail degree, is one of important indexes for evaluating the performance of the sensor and remote sensing information, and is also an important basis for identifying the shape and size of the ground object target. The spatial resolution of the remote sensing image has a direct relation with the imaging optical system, if the resolution is low, more mixed pixels exist in the remote sensing image, the analysis and understanding of the image are seriously influenced, and the target classification, detection and identification are very unfavorable.
Spectral resolution refers to the degree of fineness to which a sensor discretely samples the surface feature spectrum over a range of wavelengths. The spectral resolution is a main index for representing the performance of the sensor for acquiring the spectral information of the ground object. Compared with space image information, as another mode for scribing the feature of the ground feature, the spectral information obtained by remote detection can also realize the identification of the ground feature, and the spectral information is directly related to the material composition of the target, and is particularly more suitable for target identification, fine classification of vegetation, quantitative monitoring of marine water color, identification of camouflage in military and the like from the perspective of spectrum than the space image.
Hyperspectral Remote Sensing (Hyperspectral Remote Sensing) is a brand new Remote Sensing technology developed in the 80's of the 20 th century. The technology utilizes satellite-borne or airborne imaging spectrometer equipment to image the ground, and the imaging spectrometer forms dozens or even hundreds of narrow wave bands for continuous spectrum coverage through dispersion on each spatial pixel while imaging the spatial characteristics of a target, thereby forming remote sensing data with the spectral resolution reaching the nanometer order of magnitude. Such data is often referred to as hyperspectral data or hyperspectral images due to the high spectral resolution. The spectral resolution of the hyperspectral data is about 10 nanometers, which is tens or hundreds of times higher than that of a Multispectral (Multispectral) image. With the continuous development of imaging spectroscopy technology, hyperspectral data has been applied to many fields. The method is widely applied to environmental monitoring, urban planning, crop estimation, flood disaster investigation, homeland resource investigation, satellite reconnaissance, target detection and identification and the like in the civil field.
The hyperspectral image has the outstanding characteristic that the two-dimensional space scene information of the target image is obtained, and simultaneously, the spectral information of the physical attribute of the target image can be obtained through one-dimensional representation with high resolution, namely, the atlas is integrated. By processing the spatial features and the spectral features of the target image in the hyperspectral image, the ground object target can be distinguished and distinguished with high confidence. The method has important application significance and great potential for remote sensing image military reconnaissance, true/false target identification, agriculture and forestry fine classification and the like. Two major development trends of remote sensing technology for a long time are development towards high spatial resolution and high spectral resolution, but the development of the two is often contradictory and restricted, mainly due to the design and implementation limitations of an imaging optical system. The spectral resolution of a hyperspectral image is generally higher, but the spatial resolution of the hyperspectral image is lower, which is unfavorable for a target identification algorithm. Today, the remote sensing technology is rapidly developed, the resolution of a remote sensing image is required to be higher and higher, but the existing imaging equipment is far from meeting the requirements of all aspects due to the manufacturing process and the constraint of the existing technology. Therefore, the software method for effectively improving the spatial resolution of the hyperspectral imaging system has important practical value and practical significance.
The existing method is to carry out spatial super-resolution by a processing technology of a spatial domain or a frequency domain from the spatial geometric relation of a hyperspectral image. However, in the process, spectral information is rarely utilized, so that the algorithm is limited to the super resolution of the original general digital image, and the advantage of hyperspectrum is not fully exerted. Or only the spectral information is considered to obtain the proportion of each component in the mixed pixel, and the spatial information is rarely utilized to carry out further refinement processing. In a word, the conventional algorithm does not utilize the spectrum and the spatial information in a synergistic manner, so that the characteristics of the hyperspectrum are not fully embodied, and the defect that the information is not fully utilized exists.
Disclosure of Invention
The invention aims to solve the problem that the spatial information and the spectral information cannot be fully utilized to improve the image resolution in the existing hyperspectral image processing technology, and further provides a method for improving the resolution by the cooperation of the spatial-spectral information of a hyperspectral image.
The method comprises the following steps:
the method comprises the following steps: inputting hyperspectral image data: reading hyperspectral image data according to a data format and simultaneously inputting the hyperspectral image data into a spatial information extraction module in the step A and a spectral information extraction module in the step B of a computer for information extraction;
step A: spatial information extraction: selecting characteristic wave bands and performing spatial local analysis and judgment through a spatial information module;
step A, firstly: selecting a characteristic wave band: the characteristic wave band selection is to extract or select a characteristic spectrum band with large information amount of the hyperspectral image data;
step A two: spatial local analysis and judgment: extracting edge information of the characteristic spectrum band extracted or selected in the step A by using a differential operator edge detection algorithm, and judging edge pixels by using the edge information; non-edge pixels, interpolation processing is carried out by utilizing a copying technology to obtain space category information and correlation information, and the space category information and the correlation information are input into the step three; b, acquiring a mixed pixel, namely the edge pixel, which is the mixed pixel, and inputting the mixed pixel metadata into the step B;
and B: extracting spectral information: extracting a spectrum end member and decomposing a mixed pixel element through a spectrum information extraction module;
step B, first: spectrum end member extraction: extracting spectral end members in the hyperspectral image by using a PPI (pulse-based image) or N-FINDR (N-FidnR) algorithm, and recording spectral information and position information of the end members;
step B two: and (3) mixed pixel decomposition: acquiring the proportion of each component from the mixed pixel by using the obtained spectral information and position information of the end member through a support vector machine method;
step three: space-spectrum cooperative super resolution: distributing the proportional distribution of each component after the unmixing of the mixed pixel obtained in the step B and the spatial category information and the mutual relation information obtained by the non-edge pixel obtained in the step A according to a correlation criterion;
step four: and obtaining the hyperspectral image with improved resolution.
Compared with the prior art, the invention has the following advantages:
(1) The problem that the utilization rate of the information of the existing algorithm is low is effectively solved, and the limit of the spatial resolution during image acquisition can be broken through;
(2) The spatial resolution of the hyperspectral image is improved by utilizing a space-spectrum information cooperation technology for solving the hybrid local space correlation by using a support vector machine, the detection and positioning capabilities of the target can be greatly improved, the limitation of an image acquisition means is broken through, and the deficiency of hardware is made up.
Drawings
FIG. 1 is a flow chart of the present invention; FIG. 2 is a schematic diagram of a hybrid picture element with other neighboring picture elements; FIG. 3 is a diagram of unmixing components; FIG. 4 is a first spatial distribution diagram corresponding to an unmixed component diagram; fig. 5 is a second spatial distribution diagram corresponding to the unmixing component diagram.
Detailed Description
The first embodiment is as follows: the present embodiment is described with reference to fig. 1, and the steps of the present embodiment are as follows:
the method comprises the following steps: inputting hyperspectral image data: reading in hyperspectral image data according to a data format and simultaneously inputting the hyperspectral image data into a spatial information extraction module in the step A and a spectral information extraction module in the step B of a computer for information extraction;
step A: spatial information extraction: selecting characteristic wave bands and performing spatial local analysis and judgment through a spatial information module;
step A, firstly: selecting a characteristic wave band: the characteristic wave band selection is to extract or select a characteristic spectrum band with large information amount of the hyperspectral image data;
step A two: spatial local analysis and judgment: extracting edge information of the characteristic spectrum band extracted or selected in the step A by using a differential operator edge detection algorithm, and judging edge pixels by using the edge information; non-edge pixels, interpolation processing is carried out by utilizing a copying technology to obtain space category information and correlation information, and the space category information and the correlation information are input into the step three; b, acquiring a mixed pixel, namely the edge pixel, which is the mixed pixel, and inputting the mixed pixel metadata into the step B;
and B: extracting spectral information: extracting a spectrum end member and decomposing a mixed pixel element through a spectrum information extraction module;
step B, first: spectrum end member extraction: extracting spectral end members in the hyperspectral image by using a PPI (Point image correlation) or N-FINDR (N-Fidner) algorithm, and recording spectral information and position information of the end members;
step B two: and (3) mixed pixel decomposition: acquiring the proportion of each component from the mixed pixel by using the obtained spectral information and position information of the end member through a support vector machine method;
step three: space-spectrum cooperative super resolution: distributing the proportional distribution of each component after the unmixing of the mixed pixel obtained in the step B and the spatial category information and the mutual relation information obtained by the non-edge pixel obtained in the step A according to a correlation criterion;
step four: and obtaining the hyperspectral image with improved resolution.
The second embodiment is as follows: the difference between the present embodiment and the specific embodiment is that the method adopted for selecting the characteristic band in the step a is to extract the characteristic band through principal component transformation or select the characteristic band based on the principle of maximum entropy; other components and connection modes are the same as those of the first embodiment.
The third concrete implementation mode: the present embodiment is described with reference to fig. 2, and the difference between the present embodiment and the first embodiment lies in that the spatial local analysis and determination in the second step a is to analyze the high-spectrum image of each feature spectrum band, where x is i，J Is a pixel to be processed in a single-waveband hyperspectral image, W is a local area window taking the pixel as the center, and a window of 3 multiplied by 3 is taken as an example for explanation:
applying an edge detection algorithm in the local area window W to judge edge pixels; x is the number of i，j If the pixel is not the edge pixel, directly performing pixel copying processing on the pixel; x is the number of i，j If the pixel is an edge pixel, inputting data to the second step to carry out mixed pixel decomposition.
The local standard deviation can reflect the degree of local gray scale change, so that the smoothness of the local standard deviation can be well reflected. We use it as an edge detection operator. For the pixel x i，j And its 3 x 3 local window W, its local standard deviation sigma ij 2 Is calculated by the formula
In the formula (I), the compound is shown in the specification,
is the average value of the pixel gray levels in the window. In particular, for each pixel point x to be interpolated i，j The method for judging whether the edge point is an edge point comprises the following steps: computing a pixel x i，j The standard variance sigma of the pixel in the neighborhood window W ij 2 If and only if σ ij 2 X is considered to be x when the value of (A) exceeds a predetermined threshold value i，j And obtaining the accurate edge pixel points. Other components and connection modes are the same as those of the first embodiment.
The fourth concrete implementation mode: the first difference between this embodiment and the specific embodiment lies in that the mixed pixel decomposition in step B is performed by using a support vector machine method, and the linear spectrum mixed model can be expressed as
Where x is the observed spectral vector, s 1 ，s 2 ，...，s M Are M linearly independent spectral end members. a. a is 2 ，...，a M To respond to the mixing ratio, called "abundance", w represents additive noise. In general, the abundance a 1 ，a 2 ，...，a M Satisfy the constraints of unity and nonnegative, i.e.
In the case of two types of blending, the blending model can be rewritten as
x＝a 1 s 1 +a 2 s 2 +w＝as 1 +(1-a)s 2 +w (5)
Generally we assume w is white gaussian noise, when the least mean square estimate of x can be written as
The output of the support vector machine can be written as
Wherein x i Is a support vector, which is located near the classification boundary; y is i For the corresponding support vector x i The category is, the value is +1 or-1; alpha is alpha i Is a lagrange multiplier; k (x, x) i ) Representing an input vector x and a support vector x i The kernel function of (2) is output; b is an offset.
Equation (7) shows that for input vector x, its support vector machine output is K (x, x) i ) And these weights are known. In the process of spectrum interpretation by using the support vector machine, the relation between the output g (x) of the support vector machine and the required abundance needs to be searched. Further, we use the least mean square estimation of x
Instead of x, two types of situations are considered simultaneously, so that the current main problem is to find
Relationship to a.
From equation (6), we obtain
Wherein, l (=(s) = (a)) 1 -s 2 )+s 2 As linear transformations of a term, include translation, scale and rotation. Equation (10) shows that if the form of the kernel function is written as f (, x), thatChinese character' Tao
Can be written as a function of a variant thereof.
If a gaussian kernel function is used, then,
is also a gaussian function, and the width of the core is constant
If the kernel function and the support vector x of the two end members and i is expressed by the following formula, corresponding to a respective a =1 and a =0
And
the undetermined coefficient in equation (9) can be expressed as
thus derived a and 1-respectively, is the abundance of both mixing problems, i.e.the mixing ratio in the unmixing component diagram. Other components and connection modes are the same as those of the first embodiment.
The fifth concrete implementation mode: the present embodiment will be described with reference to fig. 3, 4 and 5, and the first difference between the present embodiment and the present embodiment lies in step three, and fig. 5 shows a stronger spatial distribution in the two spatial distribution relationships in the high-resolution image windows in fig. 4 and 5 corresponding to the spectral unmixing component map in fig. 3
Correlation, and therefore a more reasonable super-resolution result. BP neural network can be implementedBased on the nonlinear mapping from the input space to the output space, a BP network is utilized to train the functional relationship between the input and the output, namely the model is used to express the functional relationship between each pixel value in the unmixing component window and the spatial distribution of the sub-pixels in the central pixel. Let window W contain 3 × 3 pixels, x ij For the central pixel element, the amplification factor is z =2, and the functional relationship can be expressed by:
wherein
The model is based on the assumption of spatial correlation, i.e. that values of neighboring pixels are closer than values of distant pixels. The sub-pel value may be determined by the corresponding pel value and its neighborhood pel values. The larger the value of the neighboring image element, the greater the probability that a sub-image element is targeted, which maximizes the spatial correlation. The BP network can obtain the space distribution rule of the high-resolution image through training and store the space distribution rule in the connection weight.
Expression (12) gives a pair of learning patterns. Unmixing the component map window as an input pattern with vectors
X＝(x i-1，j-1 ，x i-1，j ，x i-1，j+1 ，x i，j-1 ，x i，j ，x i，j+1 ，x i+1，j-1 ，x i+1，j ，x i+1，j+1 ) T Is input to the network. One sub-pixel in the high resolution image is used as an output mode to
Is output in the form of (1). Therefore, expression (12) requires the establishment of 4 BP network models. It can be seen that the number of BP network input nodes is determined only by the unmixing component map window size.
After the learning mode is determined, the network begins training. A large number of samples are required to train the network and both of these samples satisfy the spatial correlation principle. Assuming that N pixels are included in the neighborhood of a sub-pixel, the spatial correlation can be measured by the following formula:
wherein the weight value w k Can be defined as the inverse of the distance from the sub-picture element to the centre of the k-th picture element or the inverse of the square of this distance, alpha k Unreasonable samples can be analyzed and eliminated by the metric for the unmixed components obtained in the step B two, and a super-resolution image with improved resolution can be obtained by pixel-by-pixel operation. Other components and connection modes are the same as those of the first embodiment.
Claims (1)
1. The method for improving the resolution ratio by the cooperation of the space-spectrum information of the hyperspectral image is characterized by comprising the following steps of:
the method comprises the following steps: inputting hyperspectral image data: reading hyperspectral image data according to a data format and simultaneously inputting the hyperspectral image data into a spatial information extraction module in the step A and a spectral information extraction module in the step B of a computer for information extraction;
step A: spatial information extraction: selecting characteristic wave bands and performing spatial local analysis and judgment through a spatial information module;
step A, firstly: selecting a characteristic wave band: the characteristic wave band selection is to extract or select a characteristic spectrum band with large information amount of the hyperspectral image data;
step A two: spatial local analysis and judgment: extracting edge information of the characteristic spectrum band extracted or selected in the step A by using a differential operator edge detection algorithm, and judging edge pixels by using the edge information; non-edge pixels, interpolation processing is carried out by utilizing a copying technology to obtain space category information and correlation information, and the space category information and the correlation information are input into the step three; b, acquiring a mixed pixel, namely the edge pixel, which is the mixed pixel, and inputting the mixed pixel metadata into the step B;
and B: extracting spectral information: extracting a spectrum end member and decomposing a mixed pixel element through a spectrum information extraction module;
step B, first: spectrum end member extraction: extracting spectral end members in the hyperspectral image by using a PPI (pulse-based image) or N-FINDR (N-FidnR) algorithm, and recording spectral information and position information of the end members;
step B two: and (3) mixed pixel decomposition: acquiring the proportion of each component from the mixed pixel by using the obtained spectral information and position information of the end member through a support vector machine method;
step three: space-spectrum cooperative super resolution: distributing the proportional distribution of each component after the unmixing of the mixed pixel obtained in the step B and the spatial category information and the mutual relation information obtained by the non-edge pixel obtained in the step A according to a correlation criterion;
step four: and obtaining the hyperspectral image with improved resolution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007101444495A CN100514085C (en) | 2007-10-16 | 2007-10-16 | Method for enhancing distinguishability cooperated with space-optical spectrum information of high optical spectrum image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007101444495A CN100514085C (en) | 2007-10-16 | 2007-10-16 | Method for enhancing distinguishability cooperated with space-optical spectrum information of high optical spectrum image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101140325A true CN101140325A (en) | 2008-03-12 |
| CN100514085C CN100514085C (en) | 2009-07-15 |
Family
ID=39192351
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2007101444495A Expired - Fee Related CN100514085C (en) | 2007-10-16 | 2007-10-16 | Method for enhancing distinguishability cooperated with space-optical spectrum information of high optical spectrum image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100514085C (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074009A (en) * | 2011-01-06 | 2011-05-25 | 哈尔滨工程大学 | Multiple endmember spectral mixture analysis method for hyper-spectral image |
| CN101702021B (en) * | 2009-11-06 | 2011-10-26 | 华中科技大学 | Method for obtaining spectrum-integrated time-varying target optical spectrum information and device thereof |
| CN101477197B (en) * | 2009-01-24 | 2011-12-21 | 哈尔滨工业大学 | Simulation method used for woodland complex scene high-spectrum remote sensing data |
| CN101540049B (en) * | 2009-04-29 | 2012-02-08 | 中国人民解放军海军装备研究院信息工程技术研究所 | End member extract method of hyperspectral image |
| CN101697008B (en) * | 2009-10-20 | 2012-07-04 | 北京航空航天大学 | Hyperspectral unmixing method for estimating regularized parameter automatically |
| CN103268502A (en) * | 2013-06-05 | 2013-08-28 | 中国科学院对地观测与数字地球科学中心 | Hyperspectral image waveband selecting method and device |
| CN103471715A (en) * | 2013-09-02 | 2013-12-25 | 北京航空航天大学 | Common optical path combined optical field spectral imaging method and device |
| CN103593379A (en) * | 2012-08-17 | 2014-02-19 | 通用电气航空***有限责任公司 | Method of selecting an algorithm for use in processing hyperspectral data |
| CN104036298A (en) * | 2013-09-23 | 2014-09-10 | 苏州工业职业技术学院 | High-spectrum remote sensing image end-member classification method based on Fisher self-adaptive learning |
| CN102117483B (en) * | 2009-12-31 | 2014-11-05 | 核工业北京地质研究院 | Fusion method of multispectral remote sensing images with different spatial resolutions |
| CN104236710A (en) * | 2014-09-29 | 2014-12-24 | 杭州彩谱科技有限公司 | Spectrum super-resolution method of handheld light source color illumination photometry meter |
| CN104574283A (en) * | 2013-10-23 | 2015-04-29 | 核工业北京地质研究院 | Hyperspectral image processing method for extracting pyrophyllite information |
| CN104732488A (en) * | 2013-12-20 | 2015-06-24 | 核工业北京地址研究院 | Hyperspectral image processing method for actinolite information extraction |
| CN105260993A (en) * | 2015-10-23 | 2016-01-20 | 哈尔滨工业大学 | Hyperspectral image stripe missing restoring method based on edge constraint and self-adaptive morphological filter |
| CN106097252A (en) * | 2016-06-23 | 2016-11-09 | 哈尔滨工业大学 | High spectrum image superpixel segmentation method based on figure Graph model |
| CN106595873A (en) * | 2017-01-03 | 2017-04-26 | 哈尔滨工业大学 | Subpixel temperature retrieval method based on long-wave infrared atmospheric bottom-layer radiation and visible light band linear mixed model |
| CN108593104A (en) * | 2018-03-30 | 2018-09-28 | 北京化工大学 | A kind of small-sized high s/n ratio hand-held spectral detection system |
| CN109447951A (en) * | 2018-09-30 | 2019-03-08 | 中国科学院西安光学精密机械研究所 | High optical spectrum image end member extraction method based on Givens rotation |
| CN109580495A (en) * | 2018-11-28 | 2019-04-05 | 重庆工商大学 | A kind of solution mixing device and method based on high spectrum image |
| CN109708754A (en) * | 2018-11-11 | 2019-05-03 | 上海卫星工程研究所 | For the EO-1 hyperion detection system and method for point target |
| CN110111276A (en) * | 2019-04-29 | 2019-08-09 | 西安理工大学 | Based on sky-spectrum information deep exploitation target in hyperspectral remotely sensed image super-resolution method |
| CN110119780A (en) * | 2019-05-10 | 2019-08-13 | 西北工业大学 | Based on the hyperspectral image super-resolution reconstruction method for generating confrontation network |
| CN111638185A (en) * | 2020-05-09 | 2020-09-08 | 哈尔滨工业大学 | Remote sensing detection method based on unmanned aerial vehicle platform |
| CN113280919A (en) * | 2021-05-08 | 2021-08-20 | 中国科学院长春光学精密机械与物理研究所 | Target object spectrum detection method, spectrum detection device and storage medium |
| CN113446998A (en) * | 2021-06-29 | 2021-09-28 | 哈尔滨工业大学 | Hyperspectral target detection data-based dynamic unmixing method |
-
2007
- 2007-10-16 CN CNB2007101444495A patent/CN100514085C/en not_active Expired - Fee Related
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101477197B (en) * | 2009-01-24 | 2011-12-21 | 哈尔滨工业大学 | Simulation method used for woodland complex scene high-spectrum remote sensing data |
| CN101540049B (en) * | 2009-04-29 | 2012-02-08 | 中国人民解放军海军装备研究院信息工程技术研究所 | End member extract method of hyperspectral image |
| CN101697008B (en) * | 2009-10-20 | 2012-07-04 | 北京航空航天大学 | Hyperspectral unmixing method for estimating regularized parameter automatically |
| CN101702021B (en) * | 2009-11-06 | 2011-10-26 | 华中科技大学 | Method for obtaining spectrum-integrated time-varying target optical spectrum information and device thereof |
| CN102117483B (en) * | 2009-12-31 | 2014-11-05 | 核工业北京地质研究院 | Fusion method of multispectral remote sensing images with different spatial resolutions |
| CN102074009B (en) * | 2011-01-06 | 2012-07-11 | 哈尔滨工程大学 | Multiple endmember spectral mixture analysis method for hyper-spectral image |
| CN102074009A (en) * | 2011-01-06 | 2011-05-25 | 哈尔滨工程大学 | Multiple endmember spectral mixture analysis method for hyper-spectral image |
| CN103593379A (en) * | 2012-08-17 | 2014-02-19 | 通用电气航空***有限责任公司 | Method of selecting an algorithm for use in processing hyperspectral data |
| CN103593379B (en) * | 2012-08-17 | 2018-10-19 | 通用电气航空***有限责任公司 | The method for selecting the algorithm for being used in handling high-spectral data |
| CN103268502A (en) * | 2013-06-05 | 2013-08-28 | 中国科学院对地观测与数字地球科学中心 | Hyperspectral image waveband selecting method and device |
| CN103268502B (en) * | 2013-06-05 | 2016-04-20 | 中国科学院对地观测与数字地球科学中心 | Hyperspectral image band selection method and device |
| CN103471715B (en) * | 2013-09-02 | 2015-09-09 | 北京航空航天大学 | A kind of light path combined type light field spectrum imaging method and device altogether |
| CN103471715A (en) * | 2013-09-02 | 2013-12-25 | 北京航空航天大学 | Common optical path combined optical field spectral imaging method and device |
| CN104036298A (en) * | 2013-09-23 | 2014-09-10 | 苏州工业职业技术学院 | High-spectrum remote sensing image end-member classification method based on Fisher self-adaptive learning |
| CN104574283A (en) * | 2013-10-23 | 2015-04-29 | 核工业北京地质研究院 | Hyperspectral image processing method for extracting pyrophyllite information |
| CN104732488A (en) * | 2013-12-20 | 2015-06-24 | 核工业北京地址研究院 | Hyperspectral image processing method for actinolite information extraction |
| CN104732488B (en) * | 2013-12-20 | 2018-07-06 | 核工业北京地质研究院 | A kind of Technique for Hyper-spectral Images Classification for actinolite information extraction |
| CN104236710B (en) * | 2014-09-29 | 2015-12-02 | 杭州彩谱科技有限公司 | A kind of spectrum ultra-resolution method of hand-hold light source color illumination photometry instrument |
| CN104236710A (en) * | 2014-09-29 | 2014-12-24 | 杭州彩谱科技有限公司 | Spectrum super-resolution method of handheld light source color illumination photometry meter |
| CN105260993A (en) * | 2015-10-23 | 2016-01-20 | 哈尔滨工业大学 | Hyperspectral image stripe missing restoring method based on edge constraint and self-adaptive morphological filter |
| CN106097252A (en) * | 2016-06-23 | 2016-11-09 | 哈尔滨工业大学 | High spectrum image superpixel segmentation method based on figure Graph model |
| CN106097252B (en) * | 2016-06-23 | 2019-03-12 | 哈尔滨工业大学 | High spectrum image superpixel segmentation method based on figure Graph model |
| CN106595873A (en) * | 2017-01-03 | 2017-04-26 | 哈尔滨工业大学 | Subpixel temperature retrieval method based on long-wave infrared atmospheric bottom-layer radiation and visible light band linear mixed model |
| CN106595873B (en) * | 2017-01-03 | 2019-03-29 | 哈尔滨工业大学 | Sub-pixed mapping temperature retrieval method based on LONG WAVE INFRARED surface air radiation and visible light wave range linear mixed model |
| CN108593104B (en) * | 2018-03-30 | 2020-06-26 | 北京化工大学 | Small-size high SNR hand-held type spectrum detecting system |
| CN108593104A (en) * | 2018-03-30 | 2018-09-28 | 北京化工大学 | A kind of small-sized high s/n ratio hand-held spectral detection system |
| CN109447951B (en) * | 2018-09-30 | 2021-06-22 | 中国科学院西安光学精密机械研究所 | Hyperspectral image end member extraction method based on Givens rotation |
| CN109447951A (en) * | 2018-09-30 | 2019-03-08 | 中国科学院西安光学精密机械研究所 | High optical spectrum image end member extraction method based on Givens rotation |
| CN109708754A (en) * | 2018-11-11 | 2019-05-03 | 上海卫星工程研究所 | For the EO-1 hyperion detection system and method for point target |
| CN109580495A (en) * | 2018-11-28 | 2019-04-05 | 重庆工商大学 | A kind of solution mixing device and method based on high spectrum image |
| CN110111276A (en) * | 2019-04-29 | 2019-08-09 | 西安理工大学 | Based on sky-spectrum information deep exploitation target in hyperspectral remotely sensed image super-resolution method |
| CN110111276B (en) * | 2019-04-29 | 2021-09-10 | 西安理工大学 | Hyperspectral remote sensing image super-resolution method based on space-spectrum information deep utilization |
| CN110119780A (en) * | 2019-05-10 | 2019-08-13 | 西北工业大学 | Based on the hyperspectral image super-resolution reconstruction method for generating confrontation network |
| CN110119780B (en) * | 2019-05-10 | 2020-11-27 | 西北工业大学 | Hyper-spectral image super-resolution reconstruction method based on generation countermeasure network |
| CN111638185A (en) * | 2020-05-09 | 2020-09-08 | 哈尔滨工业大学 | Remote sensing detection method based on unmanned aerial vehicle platform |
| CN111638185B (en) * | 2020-05-09 | 2022-05-17 | 哈尔滨工业大学 | Remote sensing detection method based on unmanned aerial vehicle platform |
| CN113280919A (en) * | 2021-05-08 | 2021-08-20 | 中国科学院长春光学精密机械与物理研究所 | Target object spectrum detection method, spectrum detection device and storage medium |
| CN113446998A (en) * | 2021-06-29 | 2021-09-28 | 哈尔滨工业大学 | Hyperspectral target detection data-based dynamic unmixing method |
| CN113446998B (en) * | 2021-06-29 | 2022-09-30 | 哈尔滨工业大学 | Hyperspectral target detection data-based dynamic unmixing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100514085C (en) | 2009-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101140325A (en) | Method for enhancing distinguishability cooperated with space-optical spectrum information of high optical spectrum image | |
| Wang et al. | Fusion of Landsat 8 OLI and Sentinel-2 MSI data | |
| Xie et al. | Hyperspectral image super-resolution using deep feature matrix factorization | |
| Song et al. | Spatiotemporal satellite image fusion through one-pair image learning | |
| Im et al. | A change detection model based on neighborhood correlation image analysis and decision tree classification | |
| Kotwal et al. | A novel approach to quantitative evaluation of hyperspectral image fusion techniques | |
| Prasad et al. | Optical remote sensing | |
| Li et al. | Hyperspectral pansharpening via improved PCA approach and optimal weighted fusion strategy | |
| Xu et al. | Hyperspectral image denoising using a spatial–spectral monte carlo sampling approach | |
| Qin et al. | Spatiotemporal inferences for use in building detection using series of very-high-resolution space-borne stereo images | |
| Markham et al. | A review of methods for scaling remotely sensed data for spatial pattern analysis | |
| Xie et al. | Trainable spectral difference learning with spatial starting for hyperspectral image denoising | |
| Rangzan et al. | Supervised cross-fusion method: a new triplet approach to fuse thermal, radar, and optical satellite data for land use classification | |
| Su et al. | Superpixel-based weighted collaborative sparse regression and reweighted low-rank representation for hyperspectral image unmixing | |
| Licciardi et al. | Image fusion and spectral unmixing of hyperspectral images for spatial improvement of classification maps | |
| Meng et al. | Improving the spatial resolution of hyperspectral image using panchromatic and multispectral images: An integrated method | |
| Hashim et al. | Geometric and radiometric evaluation of RazakSAT medium-sized aperture camera data | |
| Meenakshisundaram | Quality assessment of IKONOS and Quickbird fused images for urban mapping | |
| CN115100075A (en) | Hyperspectral panchromatic sharpening method based on spectral constraint and residual error attention network | |
| Toosi et al. | Object-based spectral quality assessment of high-resolution pan-sharpened satellite imageries: new combined fusion strategy to increase the spectral quality | |
| Samadzadegan et al. | Evaluating the sensitivity of image fusion quality metrics to image degradation in satellite imagery | |
| Tateishi et al. | Study on the spectral quality preservation derived from multisensor image fusion techniques between JERS-1 SAR and Landsat TM data | |
| Khandelwal et al. | Hyperspectral image enhancement based on sensor simulation and vector decomposition | |
| Hallabia et al. | A context-driven pansharpening method using superpixel based texture analysis | |
| Rajabi et al. | Fusion of hyperspectral and panchromatic images using spectral uumixing results |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090715 Termination date: 20111016 |