CN111899257A - Ground object spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition - Google Patents
Ground object spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition Download PDFInfo
- Publication number
- CN111899257A CN111899257A CN202010818747.3A CN202010818747A CN111899257A CN 111899257 A CN111899257 A CN 111899257A CN 202010818747 A CN202010818747 A CN 202010818747A CN 111899257 A CN111899257 A CN 111899257A
- Authority
- CN
- China
- Prior art keywords
- image
- remote sensing
- reflectivity
- temporal
- decomposition
- 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
- 238000002310 reflectometry Methods 0.000 title claims abstract description 96
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 53
- 238000000605 extraction Methods 0.000 title claims abstract description 29
- 230000003595 spectral effect Effects 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000005457 optimization Methods 0.000 claims abstract description 17
- 238000009877 rendering Methods 0.000 claims description 18
- 239000000126 substance Substances 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 2
- 238000012937 correction Methods 0.000 description 4
- 230000002123 temporal effect Effects 0.000 description 4
- 230000004075 alteration Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002969 morbid Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a ground object spectral reflectance image extraction method based on multi-temporal intrinsic image decomposition, and relates to multi/hyperspectral remote sensing image reflectance information extraction. The invention aims to solve the problems of low extraction precision and poor robustness of intrinsic reflectivity information of the existing multi/hyperspectral remote sensing image. The process is as follows: firstly, the method comprises the following steps: establishing a multi-temporal remote sensing image intrinsic reflectivity information expression model; II, secondly: constructing a decomposition constraint model under the constraint of local time-space energy; thirdly, the method comprises the following steps: converting the optimal solution of the decomposition constraint model into an iterative optimization solution model of the common reflectivity and the 2 shadow components; fourthly, the method comprises the following steps: giving a multi-temporal remote sensing image and initialization parameters, and solving a final intrinsic reflectivity based on the iterative optimization solution model in the third step. The invention is used in the field of digital image processing.
Description
Technical Field
The invention relates to multi/hyperspectral remote sensing image reflectivity information extraction.
Background
The spectral reflectivity information of the surface features is a main means (such as normalized vegetation index and normalized water body index) for realizing quantitative inversion of the remote sensing image, and is also an important way for carrying out surface feature identification and interpretation by utilizing the remote sensing image. However, the imaging process of the multi/hyperspectral remote sensing image is very complicated, and the image contains information such as light and shade components and light illumination components caused by the fluctuation distribution of the ground object space besides the information of the spectral reflectivity of the ground object. Therefore, the inversion of the spectral reflectivity information from the image is a typical ill-resolved problem, and the reliability of the extracted reflectivity is not high. In addition, different images have different imaging factors (solar altitude and atmospheric environment), so that the reflectivity extraction under different images is not stable enough, and the multi/hyperspectral remote sensing application is seriously influenced.
At present, methods for inverting the spectral reflectivity information of ground objects from multi/hyperspectral remote sensing images mainly comprise two methods, namely relative spectral correction and reflectivity information inversion based on intrinsic image decomposition. Since the relative spectral correction requires that the reference image must be the reflectivity information extracted after strict correction, the application range is small. Intrinsic image decomposition is a typical ill-conditioned decomposition problem, and robustness and reflectivity extraction accuracy are insufficient.
In order to improve the extraction precision of the ground object reflectivity of the multi/hyperspectral remote sensing image based on the eigen image decomposition, the method can be started by increasing the prior parameters input in the eigen decomposition. The quantity of the input remote sensing images is increased, namely, the multi-temporal intrinsic image decomposition method is constructed, so that the reflectivity extraction accuracy and robustness can be improved obviously in theory.
Disclosure of Invention
The invention aims to solve the problems of low extraction precision and poor robustness of intrinsic reflectivity information of a multi/hyperspectral remote sensing image in the prior art, and provides a ground feature spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition.
The method for extracting the spectral reflectivity image of the ground object based on the multi-temporal intrinsic image decomposition comprises the following specific processes:
the method comprises the following steps: establishing a multi-temporal remote sensing image intrinsic reflectivity information expression model, wherein the multi-temporal remote sensing image consists of the same reflectivity component and an independent shading component;
step two: constructing a decomposition constraint model under local time-space energy constraint based on a multi-temporal remote sensing image intrinsic reflectivity information expression model, obtaining an optimal solution of the decomposition constraint model by minimizing the decomposition constraint model under the local time-space energy constraint, and inverting the common reflectivity R and the shading components of each image;
step three: converting the optimal solution of the decomposition constraint model into an iterative optimization solution model of the common reflectivity and the 2 shadow components;
step four: giving a multi-temporal remote sensing image and initialization parameters, and solving a final intrinsic reflectivity based on the iterative optimization solution model in the third step;
the initialization parameters are a local neighborhood scale range, neighborhood similarity regulation and control parameters, an adjacent iteration parameter error threshold value, a reflectivity R and initial values of all image rendering components;
the initial value of each image rendering component is set to an all-zero matrix.
The invention has the beneficial effects that:
the invention aims to improve the extraction precision and stability of intrinsic reflectivity information of a multi/hyperspectral remote sensing image, and provides a ground object spectral reflectivity image accurate extraction method based on multi-temporal intrinsic image decomposition. The method constructs a multi-temporal intrinsic information expression and time-space joint optimization and inversion mechanism by increasing more image prior information, and improves the accuracy and robustness of information extraction.
The method expands the traditional single multi/hyperspectral image intrinsic reflectivity extraction model into a multi-temporal (namely multi-image) mode, starts with the increase of prior information during the decomposition of the intrinsic reflectivity, and realizes the high-precision robust ground object reflectivity extraction by constructing a local space-time energy constraint model and a corresponding iterative solution method.
In order to verify the performance of the invention, verification is carried out on a group of real high-resolution No. 2 multi-temporal remote sensing images, and experimental results show that compared with a single image reflectivity extraction result, the reflectivity images extracted by multi-temporal eigen decomposition are closer to the original image color, no chromatic aberration occurs, and the spectral reflectivity difference of similar ground objects is very small. The experimental result verifies the effectiveness of the ground feature spectral reflectivity image accurate extraction method based on multi-temporal intrinsic image decomposition.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2a is a high-resolution No. 2 time phase 1 remote sensing image (only RGB wave bands are displayed);
FIG. 2b is a high-resolution No. 2 time phase 2 remote sensing image (only RGB wave bands are displayed);
FIG. 3a is a graph of the single image reflectivity extraction results (only RGB bands are shown) for phase 1 with high score No. 2;
FIG. 3b is a graph of the single image reflectivity extraction results (only RGB bands are shown) for phase 2 with high score;
FIG. 3c is a graph of the reflectivity extraction results (showing only RGB bands) for multiple temporal image eigen-decomposition.
Detailed Description
The first embodiment is as follows: the method for extracting the spectral reflectivity image of the ground object based on the multi-temporal eigen image decomposition comprises the following specific processes:
the method comprises the following steps: establishing a multi-temporal remote sensing image intrinsic reflectivity information expression model, wherein the multi-temporal remote sensing image consists of the same reflectivity component and an independent shading component;
step two: based on a multi-temporal remote sensing image intrinsic reflectivity information expression model, constructing a decomposition constraint model under local time-space energy constraint, obtaining an optimal solution of the decomposition constraint model (generally, the solution of an extreme point of the constraint model is usually the optimal solution when an optimization function is solved) by minimizing the decomposition constraint model under the local time-space energy constraint, and inverting the common reflectivity R and the mapping component of each image to realize the reflectivity parameter inversion in the decomposition constraint model;
step three: for the minimum local time-space energy constraint decomposition model provided in the step two, according to the idea that the zero point of the first order reciprocal of differential is the optimal extreme point, the optimal solution of the decomposition constraint model is converted into an iterative optimization solution model (the reflectivity R of the multi-temporal remote sensing image and the reflectivity R of the temporal 1 remote sensing image I) of common reflectivity and 2 shadow components1Rendering component S1Time phase 2 remote sensing image I2Rendering component S2)；
Step four: giving a multi-temporal remote sensing image (image 1 and image 2) and initialization parameters, solving a model based on the iterative optimization of the step three, and solving the final intrinsic reflectivity (reflectivity R of the multi-temporal remote sensing image and a time-phase 1 remote sensing image I)1Rendering component S1Time phase 2 remote sensing image I2Rendering component S2)；
The initialization parameters are a local neighborhood scale range (set manually), neighborhood similarity regulation parameters, adjacent iteration parameter error thresholds, reflectivity R and initial values of each image rendering component;
the initial value of each image rendering component is set to an all-zero matrix.
The second embodiment is as follows: the first difference between the present embodiment and the specific embodiment is: establishing a multi-temporal remote sensing image intrinsic reflectivity information expression model in the first step; the specific process is as follows:
the method comprises the following steps of collecting multi-temporal multi/high remote sensing images (only two conditions of the two temporal remote sensing images are used for displaying) of the same place at different moments, wherein an intrinsic reflectivity information expression model of the multi-temporal remote sensing images is as follows:
I1＝S1R
I2＝S2R
wherein, I1For time phase 1 remote sensing image, I2Is a time phase 2 remote sensing image, R is the common reflectivity of the multi-time phase remote sensing image, I1、I2And R is m rows and n columns of B wave bands, S1As a time phase 1 remote sensing image I1A rendering component of S2As a time phase 2 remote sensing image I2A rendering component of S1And S2All m rows and n columns (size is different from the image size, since all bands share a shading component); and B is the spectrum wave band number of the multi/hyperspectral remote sensing image, and the general range is 4 to thousands.
Other steps and parameters are the same as those in the first embodiment.
The third concrete implementation mode: the present embodiment differs from the first or second embodiment in that: and in the second step, a decomposition constraint model under the constraint of local time-space energy is expressed as follows:
in order to realize the inversion and extraction of the reflectivity of the multi-temporal intrinsic image, the above model is still a morbid problem model (known as I)1And I2To obtain R, S1And S2) To resolve the above-mentioned pathological problems, ITwo prior conditions are given to construct a constraint model, wherein firstly, the similarity of the energy of the reflectivity in a local range is realized, and secondly, the proposed reflectivity and the shading component of each image must obey the information characterization model described in the second embodiment; the decomposition constraint model under the local spatiotemporal energy constraint described above can be expressed in the form:
wherein the content of the first and second substances,
e represents the local time-space energy; p is all pixels in the image, the number of which is mxn, n (i) represents pixels in a local neighborhood around the pixel point i (local neighborhood means 3x3, 5x5, 7x7 or 9x 9);
the first term on the right of the middle sign in the above formula is local energy similarity constraint, the second and third terms are signal model representation relation constraint on each image, and w in the formulaijThe neighborhood similarity of pixel i and its neighboring point j (this is taken to be the neighboring point within the local neighborhood range only, and usually the local neighborhood is taken to be 3x3, 5x5, 7x7 or 9x9) is represented by the following formula:
wherein, IibGray value representing the ith point and the b wave band of the image I, IjbExpressing the gray value of the jth point and the jth wave band of the image I, wherein sigma is a neighborhood similarity regulation parameter;
obtaining an optimal solution of the decomposition constraint model (generally, the solution of an extreme point of the constraint model is the optimal solution when an optimization function is solved) through the minimum constraint of the decomposition constraint model under the local time-space energy constraint, and inverting the common reflectivity R and the mapping component of each image to realize the reflectivity parameter inversion in the decomposition constraint model;
namely:
wherein R isbA gray scale value of the b-th band representing the reflectance R.
Other steps and parameters are the same as those in the first or second embodiment.
The fourth concrete implementation mode: the difference between this embodiment mode and one of the first to third embodiment modes is: in the third step, the optimal solution of the decomposition constraint model is converted into an iterative optimization solving model (reflectivity R of the multi-temporal remote sensing image and the temporal 1 remote sensing image I) of common reflectivity and 2 shadow components1Rendering component S1Time phase 2 remote sensing image I2Rendering component S2) (ii) a The specific process is as follows:
in order to solve the minimum energy constraint given in the third step, R, S is obtained by adopting the idea that the zero point of the first derivative of the formula is the optimal extreme point1And S2The formula is solved iteratively for three parameters. The obtained iterative optimization solution model is as follows:
wherein the content of the first and second substances,
remote sensing image I representing time phase 11Sha of pixel point iThe reciprocal of the ding component; i is1ibRemote sensing image I representing time phase 11The gray value of the ith point and the b th wave band; r1ibThe reflectivity of the ith point and the b wave band of the remote sensing image of the time phase 1 is represented;
remote sensing image I representing time phase 22The reciprocal of the shading component of the pixel point i; i is2ibRemote sensing image I representing time phase 22The gray value of the ith point and the b th wave band; r2ibThe reflectivity of the ith point and the b wave band of the remote sensing image of the time phase 2 is represented; ribThe reflectivity of the ith point and the b wave band of the multi-temporal remote sensing image is represented; rjbThe reflectivity of a jth point and a jth wave band of the multi-temporal remote sensing image is represented; w is ajiRepresents the similarity of pixel j and its neighboring point i (3x3, 5x5, 7x7, or 9x 9); n (j) represents pixels in a local neighborhood around pixel point j (local neighborhood means 3x3, 5x5, 7x7, or 9x 9); w is ajkRepresenting the similarity of pixel j and its neighborhood k; rkbAnd the reflectivity of the kth point and the b wave band of the multi-temporal remote sensing image is shown.
The fifth concrete implementation mode: the difference between this embodiment and one of the first to fourth embodiments is: giving a multi-temporal remote sensing image (image 1 and image 2) and initialization parameters in the fourth step, and solving final intrinsic reflectivity (reflectivity R of the multi-temporal remote sensing image and a time-phase 1 remote sensing image I) based on the iterative optimization solution model in the third step1Rendering component S1Time phase 2 remote sensing image I2Rendering component S2)。
R, S are obtained through setting corresponding threshold parameters and iteration end setting and through iterative solution1And S2Final inversion results of the three parameters;
the specific process is as follows:
the local neighborhood scale setting range (the selection range of local neighborhood similarity points) is [3,5,7,9 ]]The larger the numerical value is, the higher the smoothness is, and if no special setting is made, the local neighborhood scale is recommended to be selected by 5; the setting range of the neighborhood similarity control parameter sigma is 0.1-10, the larger the value is, the larger the similarity distribution weight contrast is, and when no special requirement exists, the setting range is suggested to be 1. Phase (C)The adjacent iteration parameter error threshold is 10-5；
The iteration end conditions are as follows:
where k' is the number of iterations.
Other steps and parameters are the same as in one of the first to fourth embodiments.
The sixth specific implementation mode: the difference between this embodiment and one of the first to fifth embodiments is: the local neighborhood dimension is set to 5.
Other steps and parameters are the same as those in one of the first to fifth embodiments.
The seventh embodiment: the difference between this embodiment and one of the first to sixth embodiments is: the similarity control parameter σ is set to 1.
Other steps and parameters are the same as those in one of the first to sixth embodiments.
The following examples were used to demonstrate the beneficial effects of the present invention:
the first embodiment is as follows:
the method for extracting the spectral reflectivity image of the ground object based on the multi-temporal eigen image decomposition specifically comprises the following steps:
the data used in the experiment are remote sensing images of two time phases in a high-grade 2 # major continuous region, and the image size is 450 multiplied by 450. The number of image bands is 4, and the data has undergone advance processing by atmosphere, geometric correction, and the like. Fig. 2a is a phase 1 raw data RGB band image, and fig. 2b is a phase 2 raw data RGB band image. Fig. 3a is an RGB band image of the phase 1 single image reflectivity extraction result, fig. 3b is an RGB band image of the phase 2 single image reflectivity extraction result, and fig. 3c is an RGB band image of the multi-phase remote sensing image reflectivity extraction result according to the present invention.
Compared with a single image reflectivity extraction result, the reflectivity image extracted by multi-time phase eigen decomposition is closer to the original image color, no chromatic aberration occurs, and the spectral reflectivity difference of the similar ground objects is small. The experimental result verifies the effectiveness of the ground feature spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition.
The present invention is capable of other embodiments and its several details are capable of modifications in various obvious respects, all without departing from the spirit and scope of the present invention.
Claims (7)
1. The ground feature spectral reflectivity image extraction method based on multi-temporal eigen image decomposition is characterized by comprising the following steps of: the method comprises the following specific processes:
the method comprises the following steps: establishing a multi-temporal remote sensing image intrinsic reflectivity information expression model, wherein the multi-temporal remote sensing image consists of the same reflectivity component and an independent shading component;
step two: constructing a decomposition constraint model under local time-space energy constraint based on a multi-temporal remote sensing image intrinsic reflectivity information expression model, obtaining an optimal solution of the decomposition constraint model by minimizing the decomposition constraint model under the local time-space energy constraint, and inverting the common reflectivity R and the shading components of each image;
step three: converting the optimal solution of the decomposition constraint model into an iterative optimization solution model of the common reflectivity and the 2 shadow components;
step four: giving a multi-temporal remote sensing image and initialization parameters, and solving a final intrinsic reflectivity based on the iterative optimization solution model in the third step;
the initialization parameters are a local neighborhood scale range, neighborhood similarity regulation and control parameters, an adjacent iteration parameter error threshold value, a reflectivity R and initial values of all image rendering components;
the initial value of each image rendering component is set to an all-zero matrix.
2. The method for extracting the spectral reflectance image of the surface feature based on the multi-temporal eigen image decomposition as claimed in claim 1, wherein: establishing a multi-temporal remote sensing image intrinsic reflectivity information expression model in the first step; the specific process is as follows:
collecting multi-temporal multi/high remote sensing images of the same place at different moments, wherein an intrinsic reflectivity information expression model of the multi-temporal remote sensing images is as follows:
I1＝S1R
I2＝S2R
wherein, I1For time phase 1 remote sensing image, I2Is a time phase 2 remote sensing image, R is the common reflectivity of the multi-time phase remote sensing image, I1、I2And R is m rows and n columns of B wave bands, S1As a time phase 1 remote sensing image I1A rendering component of S2As a time phase 2 remote sensing image I2A rendering component of S1And S2The sizes are m rows and n columns; and B is the spectrum wave band number of the multi/hyperspectral remote sensing image.
3. The method for extracting the spectral reflectance image of the ground feature based on the multi-temporal eigen image decomposition as claimed in claim 1 or 2, wherein: and in the second step, a decomposition constraint model under the constraint of local time-space energy is expressed as follows:
wherein the content of the first and second substances,
e represents the local time-space energy; p is all pixels in the image, the number of the pixels is mxn, N (i) represents the pixels in the local neighborhood around the pixel point i;
in the formula wijAnd representing the neighborhood similarity of the pixel i and the adjacent point j thereof, wherein the neighborhood similarity calculation formula is as follows:
wherein, IibGray value representing the ith point and the b wave band of the image I, IjbRepresenting the jth point of the image Ib, the gray value of the wave band, sigma is a neighborhood similarity regulation parameter;
obtaining the optimal solution of the decomposition constraint model through the minimum constraint of the decomposition constraint model under the local time-space energy constraint, and inverting the common reflectivity R and the shading components of each image;
namely:
wherein R isbA gray scale value of the b-th band representing the reflectance R.
4. The method for extracting the spectral reflectance image of the surface feature based on the multi-temporal eigen image decomposition as claimed in claim 3, wherein: in the third step, the optimal solution of the decomposition constraint model is converted into an iterative optimization solution model of the common reflectivity and the 2 shadow components; the specific process is as follows:
the iterative optimization solution model is as follows:
wherein the content of the first and second substances,
remote sensing image I representing time phase 11The reciprocal of the shading component of the pixel point i; i is1ibRemote sensing image I representing time phase 11The gray value of the ith point and the b th wave band; r1ibThe reflectivity of the ith point and the b wave band of the remote sensing image of the time phase 1 is represented;
remote sensing image I representing time phase 22The reciprocal of the shading component of the pixel point i; i is2ibRemote sensing image I representing time phase 22The gray value of the ith point and the b th wave band; r2ibThe reflectivity of the ith point and the b wave band of the remote sensing image of the time phase 2 is represented; ribThe reflectivity of the ith point and the b wave band of the multi-temporal remote sensing image is represented; rjbThe reflectivity of a jth point and a jth wave band of the multi-temporal remote sensing image is represented; w is ajiRepresenting the similarity of the pixel j and its neighborhood i; n (j) represents pixels in a local neighborhood around pixel point j; w is ajkRepresenting the similarity of pixel j and its neighborhood k; rkbAnd the reflectivity of the kth point and the b wave band of the multi-temporal remote sensing image is shown.
5. The method for extracting the spectral reflectance image of the surface feature based on the multi-temporal eigen image decomposition as claimed in claim 4, wherein: giving a multi-temporal remote sensing image and initialization parameters in the fourth step, solving a model based on the iterative optimization of the third step, and solving the final intrinsic reflectivity; the specific process is as follows:
the local neighborhood scale is set to be in the range of [3,5,7,9 ]]The setting range of the neighborhood similarity control parameter sigma is 0.1-10, and the error threshold of the adjacent iteration parameter is 10-5；
The iteration end conditions are as follows:
where k' is the number of iterations.
6. The method for extracting the spectral reflectance image of the surface feature based on the multi-temporal eigen image decomposition as claimed in claim 5, wherein: the local neighborhood dimension is set to 5.
7. The method for extracting the spectral reflectance image of the surface feature based on the multi-temporal eigen image decomposition as claimed in claim 6, wherein: the similarity control parameter σ is set to 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010818747.3A CN111899257A (en) | 2020-08-14 | 2020-08-14 | Ground object spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010818747.3A CN111899257A (en) | 2020-08-14 | 2020-08-14 | Ground object spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111899257A true CN111899257A (en) | 2020-11-06 |
Family
ID=73230280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010818747.3A Pending CN111899257A (en) | 2020-08-14 | 2020-08-14 | Ground object spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111899257A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112784747A (en) * | 2021-01-22 | 2021-05-11 | 哈尔滨工业大学 | Multi-scale eigen decomposition method for hyperspectral remote sensing image |
| CN113627357A (en) * | 2021-08-13 | 2021-11-09 | 哈尔滨工业大学 | High-spatial-high-spectral-resolution intrinsic decomposition method and system for remote sensing image |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120183213A1 (en) * | 2009-09-03 | 2012-07-19 | National Ict Australia Limited | Illumination spectrum recovery |
| CN103198483A (en) * | 2013-04-07 | 2013-07-10 | 西安电子科技大学 | Multiple time phase remote sensing image registration method based on edge and spectral reflectivity curve |
| CN111046613A (en) * | 2019-02-21 | 2020-04-21 | 北京大学 | Optimal river channel calculation method based on path tracking and river network extraction method based on multi-temporal remote sensing image |
| CN111353937A (en) * | 2020-02-28 | 2020-06-30 | 南京航空航天大学 | Super-resolution reconstruction method of remote sensing image |
-
2020
- 2020-08-14 CN CN202010818747.3A patent/CN111899257A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120183213A1 (en) * | 2009-09-03 | 2012-07-19 | National Ict Australia Limited | Illumination spectrum recovery |
| CN103198483A (en) * | 2013-04-07 | 2013-07-10 | 西安电子科技大学 | Multiple time phase remote sensing image registration method based on edge and spectral reflectivity curve |
| CN111046613A (en) * | 2019-02-21 | 2020-04-21 | 北京大学 | Optimal river channel calculation method based on path tracking and river network extraction method based on multi-temporal remote sensing image |
| CN111353937A (en) * | 2020-02-28 | 2020-06-30 | 南京航空航天大学 | Super-resolution reconstruction method of remote sensing image |
Non-Patent Citations (1)
| Title |
|---|
| 吕晓凯等: "一种基于物理机理的高光谱遥感图像真彩色校正模型", 遥感技术与应用, no. 03, 15 June 2013 (2013-06-15) * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112784747A (en) * | 2021-01-22 | 2021-05-11 | 哈尔滨工业大学 | Multi-scale eigen decomposition method for hyperspectral remote sensing image |
| CN112784747B (en) * | 2021-01-22 | 2022-03-04 | 哈尔滨工业大学 | Multi-scale eigen decomposition method for hyperspectral remote sensing image |
| CN113627357A (en) * | 2021-08-13 | 2021-11-09 | 哈尔滨工业大学 | High-spatial-high-spectral-resolution intrinsic decomposition method and system for remote sensing image |
| CN113627357B (en) * | 2021-08-13 | 2022-07-08 | 哈尔滨工业大学 | High-spatial-high-spectral-resolution intrinsic decomposition method and system for remote sensing image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | PolSAR image semantic segmentation based on deep transfer learning—Realizing smooth classification with small training sets | |
| CN108830796B (en) | Hyperspectral image super-resolution reconstruction method based on spectral-spatial combination and gradient domain loss | |
| JP6012408B2 (en) | Pan-sharpening panchromatic and multispectral images using dictionaries | |
| CN109102469B (en) | Remote sensing image panchromatic sharpening method based on convolutional neural network | |
| Xu et al. | Attention mechanism-based generative adversarial networks for cloud removal in Landsat images | |
| CN107316309B (en) | Hyperspectral image saliency target detection method based on matrix decomposition | |
| US8699790B2 (en) | Method for pan-sharpening panchromatic and multispectral images using wavelet dictionaries | |
| CN103020912B (en) | The remote sensing image restored method of a kind of combination wave band cluster and sparse expression | |
| CN113767417A (en) | Depth residual network for color filter array image denoising | |
| CN111899257A (en) | Ground object spectral reflectivity image extraction method based on multi-temporal intrinsic image decomposition | |
| CN114119444A (en) | Multi-source remote sensing image fusion method based on deep neural network | |
| CN110111276B (en) | Hyperspectral remote sensing image super-resolution method based on space-spectrum information deep utilization | |
| CN112215267B (en) | Hyperspectral image-oriented depth space spectrum subspace clustering method | |
| CN111583330B (en) | Multi-scale space-time Markov remote sensing image sub-pixel positioning method and system | |
| CN115620128A (en) | Hyperspectral anomaly detection method | |
| CN109859153B (en) | Multispectral image fusion method based on adaptive spectrum-spatial gradient sparse regularization | |
| CN113850216B (en) | Class-specific multi-mode joint representation method for large-scene remote sensing image classification | |
| CN113705538A (en) | High-resolution remote sensing image road change detection device and method based on deep learning | |
| CN114529769A (en) | Separable multi-mode joint representation method for large-scene remote sensing image classification | |
| Hsu et al. | Hyperspectral image classification via joint sparse representation | |
| CN107832790B (en) | Semi-supervised hyperspectral image classification method based on local low-rank representation | |
| CN111523451A (en) | Method for constructing high-space-time resolution NDVI data | |
| CN115082533B (en) | Near space remote sensing image registration method based on self-supervision | |
| CN114972122A (en) | Hyperspectral remote sensing image bad pixel restoration method and device, electronic equipment and medium | |
| Nair et al. | Hyperspectral image fusion using fast high-dimensional denoising |
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 |