CN113537150A - Hyperspectral image target anomaly detection method, system, terminal and storage medium - Google Patents
Hyperspectral image target anomaly detection method, system, terminal and storage medium Download PDFInfo
- Publication number
- CN113537150A CN113537150A CN202110921317.9A CN202110921317A CN113537150A CN 113537150 A CN113537150 A CN 113537150A CN 202110921317 A CN202110921317 A CN 202110921317A CN 113537150 A CN113537150 A CN 113537150A
- Authority
- CN
- China
- Prior art keywords
- hyperspectral image
- detection method
- representation
- image target
- weight
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 70
- 230000002159 abnormal effect Effects 0.000 claims abstract description 27
- 230000006870 function Effects 0.000 claims abstract description 15
- 230000005856 abnormality Effects 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 15
- 238000000034 method Methods 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 9
- 150000001875 compounds Chemical class 0.000 claims description 6
- 238000001228 spectrum Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 4
- 230000009977 dual effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000008407 joint function Effects 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 241001466077 Salina Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- 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
- G06T2207/10036—Multispectral image; Hyperspectral image
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Image Analysis (AREA)
Abstract
A hyperspectral image target anomaly detection method, a hyperspectral image target anomaly detection system, a terminal and a storage medium are disclosed, wherein the detection method comprises the following steps: giving each wave band of the hyperspectral image a proper weight through maximum cross-correlation entropy cooperation representation; and for the hyperspectral image with each wave band having a proper weight, obtaining a pixel abnormal score through maximum cross-correlation entropy cooperation representation, and selecting an abnormal pixel through setting a proper threshold value. According to the hyperspectral image detection method, different weights are given to different wave band self-adaptations of the hyperspectral image, so that the influence of wave band noise and abnormal pixel point segments on the detection accuracy can be effectively avoided, and the robustness of the algorithm is improved. The objective function is constructed as a joint function, the wave band weight and the neighborhood pixel point weight are solved simultaneously, an iterative solution algorithm is used for efficient solution, and the problem of suboptimal solution possibly caused by independent learning of one of the wave band weight and the neighborhood pixel point weight is solved. The invention can effectively improve the accuracy.
Description
Technical Field
The invention belongs to the field of remote sensing image processing, and particularly relates to a hyperspectral image target anomaly detection method, a hyperspectral image target anomaly detection system, a hyperspectral image target anomaly detection terminal and a storage medium.
Background
The hyperspectral image has a plurality of spectral domains, can provide a large amount of spectral information and spatial information of the earth surface, and is widely applied to the fields of scene classification, mixing, clustering, change detection, target detection and the like due to the good detection performance of the hyperspectral image. One of applications of hyperspectral images, hyperspectral anomaly detection, is one of the hotspots of current research, and currently, hyperspectral anomaly detection is widely applied to military and civil fields and achieves good effects. Hyperspectral anomaly detection is an unsupervised binary problem, and does not require prior knowledge. In the hyperspectral anomaly detection, an object whose spectrum is significantly different from its surrounding spectrum is regarded as an anomaly, and an object other than the anomaly in the entire image is defined as a background.
Many existing methods can complete hyperspectral anomaly detection. The Reed-Xiaoli (RX) algorithm is based on the Mahalanobis distance between pixels and the background and assumes that multivariate Gaussian background distributions are used to identify anomalies, detailed in turn into GRX using the entire image to model the background and LRX using local dual-window modeling. Later, another scholarly proposed a non-linear version of rx (KRX) that was extended to a feature space associated with the original input space by a non-linear mapping function. Due to the problem that KRX has a target pollution background, and the improvement of an algorithm based on the problem is carried out by scholars, a robust nonlinear anomaly detection algorithm (RNAD) using robust regression analysis in a kernel space is provided. Collaborative representation based detectors (CRDs) were proposed later and widely used. CRD assumes that every pixel in the background can be approximated using its neighborhood pixels, while outlier pixels cannot. It is assumed that the background pixel is a linear combination of neighboring pixels and is minimized byNorm ofEnhancing the collaborative representation. Currently available RX-based detectors assume that the background is uniform and are modeled using a multivariate gaussian distribution. However, in practice, this assumption is often not satisfied, which results in a decrease in accuracy.
In addition, noise and abnormal interference cannot be completely shielded when the hyperspectral image is acquired, so that noise and abnormal pixel points may exist in a certain waveband in the hyperspectral image. CRD and its derivatives assume that each band has the same importance, and such assumption will be poor in robustness to noise bands, thereby reducing the effect of anomaly detection.
The related documents are:
【1】Reed I S,Yu X.Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution[J].IEEE Transactions on Acoustics,Speech,and Signal Processing,1990,38(10):1760-1770.
【2】Molero J M,Garzón E M,García I,et al.Analysis and optimizations of global and local versions of the RX algorithm for anomaly detection in hyperspectral data[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2013,6(2):801-814.
【3】Kwon H,Nasrabadi N M.Kernel RX-algorithm:Anonlinear anomaly detector for hyperspectral imagery[J].IEEE transactions on Geoscience and Remote Sensing,2005,43(2):388-397.
【4】Zhao R,Du B,Zhang L.Arobust nonlinear hyperspectral anomaly detection approach[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(4):1227-1234.
【5】Li W,Du Q.Collaborative representation for hyperspectral anomaly detection[J].IEEE Transactions on Geoscience and Remote Sensing,2014,53(3):1463-1474.
disclosure of Invention
The invention aims to solve the problems in the prior art, and provides a hyperspectral image target abnormality detection method, a hyperspectral image target abnormality detection system, a hyperspectral image target abnormality detection terminal and a hyperspectral image target abnormality detection storage medium, which can effectively avoid the influence of noise and abnormal wave bands on hyperspectral abnormality detection.
In order to achieve the purpose, the invention has the following technical scheme:
a hyperspectral image target anomaly detection method comprises the following steps:
giving each wave band of the hyperspectral image a proper weight through maximum cross-correlation entropy cooperation representation;
and for the hyperspectral image with each wave band having a proper weight, obtaining a pixel abnormal score through maximum cross-correlation entropy cooperation representation, and selecting an abnormal pixel through setting a proper threshold value.
As a preferred scheme of the target abnormality detection method of the hyperspectral image, the hyperspectral image is defined asRepresenting a hyperspectral image with n pixels, the dimension being d dimension; local approximation using sliding dual windows, the dimensions of the outer and inner windows being w, respectivelyoutAnd winIs expressed, therefore, the pixels in the neighborhood are used Wherein s ═ wout×wout-win×win。
As a preferred scheme of the target abnormality detection method for the hyperspectral image, the maximum cross-correlation entropy cooperation representation target function of the hyperspectral image is as follows:
wherein the content of the first and second substances,indicates each one usedThe weight of the neighborhood pixels, lambda is a regularization parameter;
in the formula (I), the compound is shown in the specification,the larger the representation of the difference in collaborative representation, the larger the differenceThe smaller.
As an optimal scheme of the hyperspectral image target abnormality detection method, the maximum cross-correlation entropy cooperation representation target function is reconstructed through a semi-quadratic technology to obtain the following expression:
in the formula (I), the compound is shown in the specification,is a diagonal matrix, each element on the diagonalA weight representing the band;
wkis defined as follows:
as an optimal scheme of the hyperspectral image target abnormality detection method, a coefficient corresponding to a pixel point with larger difference with a central pixel is smaller, and a diagonal regularization matrix is utilized to solve
Wherein gamma isiFor the Tikhonov regularization matrix, the expression is as follows:
as an optimal scheme of the hyperspectral image target anomaly detection method, the proposed target function is optimized in an iterative optimization mode, and unsupervised hyperspectral image target anomaly detection is completed.
The invention also provides a hyperspectral image target anomaly detection system, which comprises:
the spectrum band weighting module is used for giving a proper weight to each band of the hyperspectral image through maximum cross-correlation entropy cooperation representation;
and the abnormal pixel screening module is used for obtaining a pixel abnormal score for the hyperspectral image with a proper weight for each wave band through maximum cross-correlation entropy cooperation representation, and selecting an abnormal pixel through setting a proper threshold value.
The invention also provides terminal equipment which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor realizes the steps of the hyperspectral image target abnormality detection method when executing the computer program.
The invention further provides a computer readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the steps of the hyperspectral image target abnormality detection method.
Compared with the prior art, the invention at least has the following beneficial effects:
in the hyperspectral image target anomaly detection process, different weights are self-adaptively given to different wave bands of a hyperspectral image, a hyperspectral image with a proper weight for each wave band is obtained, pixel anomaly scores are obtained through maximum cross-correlation entropy cooperation expression, an abnormal pixel is selected through setting a proper threshold value, wave band noise and the influence of abnormal pixel point segments on detection accuracy can be effectively avoided, and the robustness of an algorithm is improved. Compared with the current mainstream hyperspectral image target anomaly detection method, the method provided by the invention has obvious advantages, and the accuracy rate on the public data set greatly leads other methods.
Furthermore, the objective function is constructed as a joint function, the wave band weight and the neighborhood pixel point weight are solved simultaneously, an iterative solution algorithm is used for efficient solution, and the problem of suboptimal solution possibly caused by independent learning of one of the wave band weight and the neighborhood pixel point weight is solved.
Drawings
FIG. 1 is a graph comparing ROC curves on the AVIRIS-I dataset for the present invention with other methods;
FIG. 2 is a graph comparing ROC curves on ABU-airport-1 data set according to the present invention and other methods;
FIG. 3 AVIRIS-I dataset pseudo-color map;
FIG. 4 is an AVIRIS-I dataset tag map;
FIG. 5 is a graph of the detection results of the AVIRIS-I data set CRD;
FIG. 6 is a graph of the AVIRIS-I data set of the present invention;
FIG. 7 AVIRIS-II dataset pseudo-color map;
FIG. 8 AVIRIS-II dataset tag map;
FIG. 9 is a graph of the results of the detection of the AVIRIS-II data set CRD;
FIG. 10 is a graph of the AVIRIS-II data set of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Suppose a hyperspectral image is defined asRepresenting a hyperspectral image with n pixels, the dimension being d dimension; local approximation using sliding dual windows, the dimensions of the outer and inner windows being w, respectivelyoutAnd winIs expressed, therefore, the pixels in the neighborhood are usedWherein s ═ wout×wout-win×win。
S1, giving each wave band of the hyperspectral image a proper weight through maximum cross-correlation entropy cooperation;
and S2, obtaining pixel abnormal scores by the cooperation of the maximum cross-correlation entropy of the hyperspectral images with each wave band and proper weight, and selecting abnormal pixels by setting proper threshold values.
The hyperspectral image target abnormality detection method provided by the invention constructs a detector based on a maximum entropy theory (MCC) provided in Information Theory Learning (ITL). ITL provides a unified approach to improve machine learning methods for outlier robustness. The MCC may obtain more information from the data to adjust and therefore may have a greater advantage in processing non-gaussian signals. Entropy introduces a new metric that equals to two points when they are close togetherAnd (4) norm. As the two points get farther apart, the metric will approachNorm, finally approachingAnd (4) norm. Thus, MCC is strongly robust against outliers.
The maximum cross-correlation entropy cooperation of the hyperspectral image represents an objective function as follows:
wherein the content of the first and second substances,representing the weight of each neighborhood pixel used, λ being the regularization parameter;
in the formula (I), the compound is shown in the specification,the larger the representation of the difference in collaborative representation, the larger the differenceThe smaller.
Reconstructing the maximum cross-correlation entropy cooperation expression target function through a semi-quadratic technology to obtain the following expression:
in the formula (I), the compound is shown in the specification,is a diagonal matrix, each element on the diagonalA weight representing the band;
wkis defined as follows:
corresponding to a small coefficient with a pixel point with a large difference with a central pixel, and solving by using a diagonal regularization matrix Wherein gamma isiFor the Tikhonov regularization matrix, the expression is as follows:
and optimizing the proposed objective function by adopting an iterative optimization mode to complete unsupervised hyperspectral image target anomaly detection.
The hyperspectral image target anomaly detection method provided by the invention can effectively avoid the influence of noise and abnormal wave bands on hyperspectral anomaly detection. The noise band is given a sufficiently small weight by the method of the invention so as not to affect the inference result, while the contributing band is given more weight so that it can provide more effective information.
The invention also provides a hyperspectral image target anomaly detection system, which comprises:
the spectrum band weighting module is used for giving a proper weight to each band of the hyperspectral image through maximum cross-correlation entropy cooperation representation;
and the abnormal pixel screening module is used for obtaining a pixel abnormal score for the hyperspectral image with a proper weight for each wave band through maximum cross-correlation entropy cooperation representation, and selecting an abnormal pixel through setting a proper threshold value.
The invention also provides terminal equipment which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor realizes the steps of the hyperspectral image target abnormality detection method when executing the computer program.
The invention further provides a computer readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the steps of the hyperspectral image target abnormality detection method.
The computer program may be divided into one or more modules/units, which are stored in the memory and executed by the processor, to complete the hyperspectral image target abnormality detection method of the invention.
The terminal can be a desktop computer, a notebook, a palm computer, a cloud server and other computing equipment, and can also be a processor and a memory. The processor may be a Central Processing Unit (CPU), other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The memory may be used to store computer programs and/or modules, and the processor may implement various functions of the hyperspectral image target abnormality detection system by running or executing the computer programs and/or modules stored in the memory and calling data stored in the memory.
Examples
A hyperspectral image target anomaly detection method comprises the following steps:
the method comprises the following steps: inputting a hyperspectral imageRegularization term parameters λ, σ and sliding window size woutAnd win。
Step two: solving the objective function, fixing Wi(giving initial value I) update of αiThe objective function will be converted into:
for the above formula, let it pair alphaiTaking the derivative of the collocated derivative to 0, one can obtain:
Step three: fixed alphaiUpdating W by a formulai。
Step four: and repeating the second step and the third step until convergence.
Step five: and moving the sliding window to process the next pixel. And carrying out the operations of the second step, the third step and the fourth step.
Step six: by passingAnd obtaining hyperspectral full-pixel abnormal scores, and setting a threshold value to obtain abnormal pixels.
Table 1 shows that the collaborative representation method based on the maximum entropy criterion compares the high spectrum abnormality detection on the public data set with the AUC value of the current mainstream method, wherein the larger the AUC value is, the better the AUC value is.
TABLE 1
Dataset | GRX | LRX | CRD | LSMAD | SWCRD | ERCRD | Ours |
AVIRIS-I | 0.9111 | 0.8584 | 0.9911 | 0.9758 | 0.9936 | 0.9678 | 0.9985 |
AVIRIS-II | 0.9403 | 0.8466 | 0.9696 | 0.9703 | 0.9798 | 0.9614 | 0.9855 |
AVIRIS-III | 0.8771 | 0.8411 | 0.9828 | 0.9318 | 0.9837 | 0.9349 | 0.9904 |
ABU- |
0.8221 | 0.8886 | 0.923 | 0.8806 | 0.9273 | 0.897 | 0.9631 |
ABU-airport-2 | 0.8404 | 0.9706 | 0.9735 | 0.9315 | 0.9692 | 0.9693 | 0.9879 |
Salinas | 0.8872 | 0.9172 | 0.9606 | 0.9351 | 0.951 | 0.9447 | 0.9643 |
FIGS. 1 and 2 are graphs comparing ROC curves on AVIRIS-I and ABU-airport-1 data sets for the current mainstream method, with the curves in the graphs being closer to the upper left corner and more effective. Fig. 3, fig. 4, fig. 5 and fig. 6 are respectively a pseudo-color chart, a label chart, a detection effect chart of CRD and a detection effect chart of the present invention of the AVIRIS-I data set. Fig. 6, 7, 8 and 9 are respectively a pseudo-color map, a label map, a CRD detection effect map and a detection effect map of the present invention of the AVIRIS-II data set, in which the brighter the pixel, the higher the possibility that the pixel is an abnormal point. Compared with the current mainstream hyperspectral image target abnormality detection method, the method provided by the invention has obvious advantages, and the accuracy on public data sets (AVIRIS-I data sets and AVIRIS-II data sets) is greatly superior to that of other methods.
The above-mentioned embodiments are only preferred embodiments of the present invention, and are not intended to limit the technical solution of the present invention, and it should be understood by those skilled in the art that the technical solution can be modified and replaced by a plurality of simple modifications and replacements without departing from the spirit and principle of the present invention, and the modifications and replacements also fall into the protection scope covered by the claims.
Claims (9)
1. A hyperspectral image target anomaly detection method is characterized by comprising the following steps:
giving each wave band of the hyperspectral image a proper weight through maximum cross-correlation entropy cooperation representation;
and for the hyperspectral image with each wave band having a proper weight, obtaining a pixel abnormal score through maximum cross-correlation entropy cooperation representation, and selecting an abnormal pixel through setting a proper threshold value.
2. The hyperspectral image target abnormality detection method according to claim 1, characterized in that: the hyperspectral image is defined asRepresenting a hyperspectral image with n pixels, the dimension being d dimension; local approximation using sliding dual windows, the dimensions of the outer and inner windows being w, respectivelyoutAnd winIs expressed, therefore, the pixels in the neighborhood are usedWherein s ═ wout×wout-win×win。
3. The hyperspectral image target abnormality detection method according to claim 1, wherein the maximum cross-correlation entropy cooperation representation target function of the hyperspectral image is as follows:
wherein the content of the first and second substances,representing the weight of each neighborhood pixel used, λ being the regularization parameter;
4. The hyperspectral image target abnormality detection method according to claim 3, characterized in that the maximum cross-correlation entropy cooperation representation target function is reconstructed by a semi-quadratic technique to obtain the following expression:
in the formula (I), the compound is shown in the specification,is a diagonal matrix, each element on the diagonalA weight representing the band;
wkis defined as follows:
5. the hyperspectral image target abnormality detection method according to claim 4, characterized in that the pixel points with larger difference from the center pixel correspond to smaller coefficients, and the diagonal regularization matrix is used for solving
Wherein riFor the Tikhonov regularization matrix, the expression is as follows:
6. the hyperspectral image target abnormality detection method according to claim 5, characterized in that: and optimizing the proposed objective function by adopting an iterative optimization mode to complete unsupervised hyperspectral image target anomaly detection.
7. A hyperspectral image target anomaly detection system is characterized by comprising:
the spectrum band weighting module is used for giving a proper weight to each band of the hyperspectral image through maximum cross-correlation entropy cooperation representation;
and the abnormal pixel screening module is used for obtaining a pixel abnormal score for the hyperspectral image with a proper weight for each wave band through maximum cross-correlation entropy cooperation representation, and selecting an abnormal pixel through setting a proper threshold value.
8. A terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: the processor, when executing the computer program, implements the steps of the hyperspectral image target abnormality detection method according to any of claims 1 to 6.
9. A computer-readable storage medium storing a computer program, characterized in that: the computer program, when being executed by a processor, realizes the steps of the hyperspectral image target abnormality detection method according to any of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110921317.9A CN113537150B (en) | 2021-08-11 | 2021-08-11 | Hyperspectral image target anomaly detection method, system, terminal and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110921317.9A CN113537150B (en) | 2021-08-11 | 2021-08-11 | Hyperspectral image target anomaly detection method, system, terminal and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113537150A true CN113537150A (en) | 2021-10-22 |
CN113537150B CN113537150B (en) | 2024-04-02 |
Family
ID=78090891
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110921317.9A Active CN113537150B (en) | 2021-08-11 | 2021-08-11 | Hyperspectral image target anomaly detection method, system, terminal and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113537150B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010132135A1 (en) * | 2009-05-14 | 2010-11-18 | Raytheon Company | Adaptive spatial-spectral processing (assp) |
CN105989597A (en) * | 2015-02-13 | 2016-10-05 | 中国科学院西安光学精密机械研究所 | Hyperspectral image anomaly target detection method based on pixel selection process |
CN109447272A (en) * | 2018-09-29 | 2019-03-08 | 西安交通大学 | A kind of extreme learning machine method based on center of maximum cross-correlation entropy criterion |
WO2019218313A1 (en) * | 2018-05-17 | 2019-11-21 | 五邑大学 | Progressive dynamic hyperspectral image classification method |
CN110705600A (en) * | 2019-09-06 | 2020-01-17 | 西安交通大学 | Cross-correlation entropy based multi-depth learning model fusion method, terminal device and readable storage medium |
CN111160310A (en) * | 2020-01-02 | 2020-05-15 | 西北工业大学 | Hyperspectral abnormal target detection method based on self-weight collaborative representation |
-
2021
- 2021-08-11 CN CN202110921317.9A patent/CN113537150B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010132135A1 (en) * | 2009-05-14 | 2010-11-18 | Raytheon Company | Adaptive spatial-spectral processing (assp) |
CN105989597A (en) * | 2015-02-13 | 2016-10-05 | 中国科学院西安光学精密机械研究所 | Hyperspectral image anomaly target detection method based on pixel selection process |
WO2019218313A1 (en) * | 2018-05-17 | 2019-11-21 | 五邑大学 | Progressive dynamic hyperspectral image classification method |
CN109447272A (en) * | 2018-09-29 | 2019-03-08 | 西安交通大学 | A kind of extreme learning machine method based on center of maximum cross-correlation entropy criterion |
CN110705600A (en) * | 2019-09-06 | 2020-01-17 | 西安交通大学 | Cross-correlation entropy based multi-depth learning model fusion method, terminal device and readable storage medium |
CN111160310A (en) * | 2020-01-02 | 2020-05-15 | 西北工业大学 | Hyperspectral abnormal target detection method based on self-weight collaborative representation |
Non-Patent Citations (2)
Title |
---|
周千;马文涛;桂冠;: "基于l_1-范数约束的递归互相关熵的稀疏系统辨识", 信号处理, no. 09 * |
朱德辉;杜博;张良培;: "波段选择协同表达高光谱异常探测算法", 遥感学报, no. 04 * |
Also Published As
Publication number | Publication date |
---|---|
CN113537150B (en) | 2024-04-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10140522B2 (en) | Fully convolutional pyramid networks for pedestrian detection | |
Baudat et al. | Feature vector selection and projection using kernels | |
Zhang et al. | An adaptive bilateral filter based framework for image denoising | |
Xu et al. | Spectral–spatial classification of hyperspectral image based on low-rank decomposition | |
Shanmugavadivu et al. | Particle swarm optimized bi-histogram equalization for contrast enhancement and brightness preservation of images | |
CN109598750B (en) | Large-scale difference image feature point matching method based on deformation space pyramid | |
Aaron et al. | Dynamic incremental k-means clustering | |
Zhang et al. | A switching median–mean filter for removal of high-density impulse noise from digital images | |
Chen et al. | Convolutional neural network for classification of solar radio spectrum | |
Tu et al. | Hyperspectral image classification by combining local binary pattern and joint sparse representation | |
Yang et al. | A new TLD target tracking method based on improved correlation filter and adaptive scale | |
Liang et al. | Multi-scale hybrid attention graph convolution neural network for remote sensing images super-resolution | |
López-Rubio et al. | The effect of noise on foreground detection algorithms | |
Zhu et al. | Robust semi-supervised Kernel-FCM algorithm incorporating local spatial information for remote sensing image classification | |
Liu et al. | Color image segmentation using nonparametric mixture models with multivariate orthogonal polynomials | |
Rajabi et al. | A Modified adaptive hysteresis smoothing approach for image denoising based on spatial domain redundancy | |
Li et al. | Joint spectral-spatial hyperspectral image classification based on hierarchical subspace switch ensemble learning algorithm | |
Bi et al. | l 2, p-norm sequential bilateral 2DPCA: a novel robust technology for underwater image classification and representation | |
Xiao et al. | Robust Tensor Low-Rank Sparse Representation with Saliency Prior for Hyperspectral Anomaly Detection | |
CN113537150A (en) | Hyperspectral image target anomaly detection method, system, terminal and storage medium | |
Wang et al. | Dynamic super-pixel normalization for robust hyperspectral image classification | |
Xu | Sparse abundance estimation with low-rank reconstruction for hyperspectral unmixing | |
Du et al. | A novel unsupervised bands selection algorithm for hyperspectral image | |
Tseng et al. | A temperature data denoising method using Laplacian matrix and Neumann series | |
Song et al. | Superpixel-based hybrid discriminative random field for fast PolSAR image classification |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |