CN116416231A - Hyperspectral anomaly detection method and device - Google Patents
Hyperspectral anomaly detection method and device Download PDFInfo
- Publication number
- CN116416231A CN116416231A CN202310324310.8A CN202310324310A CN116416231A CN 116416231 A CN116416231 A CN 116416231A CN 202310324310 A CN202310324310 A CN 202310324310A CN 116416231 A CN116416231 A CN 116416231A
- Authority
- CN
- China
- Prior art keywords
- hyperspectral
- anomaly detection
- hyperspectral image
- algorithm
- utilizing
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 60
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 60
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000001228 spectrum Methods 0.000 claims description 18
- 230000002195 synergetic effect Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 22
- 230000003595 spectral effect Effects 0.000 abstract description 11
- 230000002159 abnormal effect Effects 0.000 abstract description 7
- 230000008569 process Effects 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000000576 coating method Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000013016 damping Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007417 hierarchical cluster analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Remote Sensing (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a hyperspectral anomaly detection method, which relates to the technical field of hyperspectral image anomaly detection and comprises the following steps: the method comprises the steps of obtaining spectral data of all wave bands of a hyperspectral image, processing the spectral data of all the wave bands by utilizing a neighbor propagation clustering algorithm to obtain clustering results of the hyperspectral image, processing all the wave bands in each clustering result by utilizing a self-adaptive wave band selection algorithm to obtain an optimal wave band set which is arranged from large to small according to ABS indexes and has the ABS indexes larger than a preset threshold value, reconstructing the hyperspectral image according to the optimal wave band set, and processing the reconstructed hyperspectral image by utilizing a collaborative representation algorithm to obtain an anomaly detection target. The method can improve the accuracy of abnormal target detection under different complex hyperspectral images and improve the efficiency of abnormal target detection.
Description
Technical Field
The invention relates to the technical field of hyperspectral image anomaly detection, in particular to a hyperspectral anomaly detection method and device.
Background
The hyperspectral information of the hyperspectral image has unique advantages in the field of remote sensing image processing, but the hyperspectral information is rich, and meanwhile, the defects of overlarge data volume and information redundancy exist. The hyperspectral image anomaly detection task has the advantages that the calculated amount of the hyperspectral image anomaly detection task is too large, the operation time is too long, and meanwhile, the accuracy of anomaly detection can be reduced due to redundant spectral information.
In order to solve the problems of overlarge data size and information redundancy of hyperspectral image hyperspectral information, most of existing hyperspectral image anomaly detection is based on band selection such as a K-means clustering algorithm, a hierarchical clustering analysis method, a maximum information amount-based algorithm, an information dispersion method and a first spectrum derivative method, and the problems of low target detection accuracy and efficiency exist when the methods are independently used.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, a first aspect of the present invention proposes a hyperspectral anomaly detection method, including:
acquiring spectrum data of all wave bands of a hyperspectral image;
processing the spectrum data of all wave bands by utilizing a neighbor propagation clustering algorithm to obtain a clustering result of the hyperspectral image;
processing all wave bands in each clustering result by utilizing a self-adaptive wave band selection algorithm to obtain an optimal wave band set which is arranged from large to small according to ABS indexes and has the ABS indexes larger than a preset threshold value, and reconstructing a hyperspectral image according to the optimal wave band set;
and processing the reconstructed hyperspectral image by utilizing a collaborative representation algorithm to obtain an anomaly detection target.
The invention also provides a hyperspectral anomaly detection device based on band selection and collaborative representation, which comprises:
the data acquisition module is used for acquiring spectrum data of all wave bands of the hyperspectral image;
the clustering module is used for processing the spectrum data of all wave bands by utilizing a neighbor propagation clustering algorithm to obtain a clustering result of the hyperspectral image;
the image reconstruction module is used for processing all the wave bands in each clustering result by utilizing a self-adaptive wave band selection algorithm, obtaining an optimal wave band set which is arranged from large to small according to the ABS index and has the ABS index larger than a preset threshold value, and reconstructing a hyperspectral image according to the optimal wave band set;
and the anomaly detection module is used for processing the reconstructed hyperspectral image by utilizing the collaborative representation algorithm to obtain an anomaly detection target.
In another aspect, the present invention also provides an electronic device, including a processor and a memory, where at least one instruction, at least one program, a code set, or an instruction set is stored, where at least one instruction, at least one program, a code set, or an instruction set is loaded and executed by the processor to implement a hyperspectral anomaly detection method as in the first aspect.
In another aspect, the present invention also provides a computer readable storage medium having stored therein at least one instruction, at least one program, code set, or instruction set, the at least one instruction, at least one program, code set, or instruction set being loaded and executed by a processor to implement a hyperspectral anomaly detection method as in the first aspect.
The embodiment of the invention provides a hyperspectral anomaly detection method and device, which have the following beneficial effects compared with the prior art:
according to the spectrum correlation of the hyperspectral image, the method utilizes a neighbor propagation clustering algorithm to perform band primary clustering. And then, selecting the wave bands in each cluster by using a self-adaptive wave band selection algorithm to form an optimal hyperspectral image wave band set, reconstructing a hyperspectral image by using the optimal hyperspectral image wave band set, carrying out anomaly detection on the reconstructed hyperspectral image by using a collaborative representation algorithm, and further improving the anomaly detection accuracy by using the adjacent local space information of the hyperspectral image.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the following description will make a brief introduction to the drawings used in the description of the embodiments or the prior art. It should be apparent that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained from these drawings without inventive effort to those of ordinary skill in the art.
FIG. 1 is a flowchart of a hyperspectral anomaly detection method according to an embodiment of the present invention;
FIG. 2 is a frame diagram of a hyperspectral anomaly detection algorithm based on band selection and collaborative representation provided by an embodiment of the present invention;
FIG. 3 is a graph of an anomaly detection AUC of a hyperspectral anomaly detection algorithm based on band selection and collaborative representation provided by an embodiment of the present invention;
FIG. 4 is a comparison chart of anomaly detection results of two sets of data sets based on a band selection and synergistic representation hyperspectral anomaly detection algorithm provided by an embodiment of the present invention;
FIG. 5 is a ROC graph of anomaly detection results of two sets of data sets based on a band selection and synergistic representation hyperspectral anomaly detection algorithm provided by an embodiment of the present invention;
fig. 6 is a block diagram of a hyperspectral anomaly detection device based on band selection and collaborative representation according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The present specification provides method operational steps as described in the examples or flowcharts, but may include more or fewer operational steps based on conventional or non-inventive labor. When implemented in a real system or server product, the methods illustrated in the embodiments or figures may be performed sequentially or in parallel (e.g., in a parallel processor or multithreaded environment).
As shown in fig. 1, a hyperspectral anomaly detection method includes:
102, processing spectrum data of all wave bands by utilizing a neighbor propagation clustering algorithm to obtain a clustering result of a hyperspectral image;
in step 102, the neighbor propagation clustering algorithm (Affinity Propagation, AP) is a simple and fast clustering algorithm, and clusters are automatically performed according to the similarity between the target data, without manually setting the number of clusters in advance. Let the hyperspectral image data be { x } 1 ,x 2 ,L,x n By calculating the similarity between sample data, the attraction information matrix R and the attribution information matrix A are continuously updated, an optimal clustering center is found by combining the similarity information and attribution information, R (i, k) represents the attraction degree of the data point k to the data point i, a (i, k) represents the attribution degree of the data point i to the data point k, s (i, k) is a deflection parameter set for each data point, and the larger the value of s (i, k) is, the larger the probability that the corresponding point is selected as a class representative point is. An algorithm generally assumes that all sample points are equally likely to be selected as class representative points, i.e., all s (i, k) are set to the same value p. The algorithm is iterated by the steps of:
attraction degree r after iteration t+1 (i, k) iterating according to formula (1),
wherein a is t (i, k') represents the attribution value of other points to the point i before iteration except k, and is initially 0; s (i, k') represents the attraction of other points to i except k, i.e. all points other than i are competing for the point iWeight;
degree of attribution after iteration a t+1 (i, k) iterating according to formulas (2) and (3),
a t+1 (k,k)＝∑ i′≠k max{0,r t (i′,k)}，i＝k (3)
wherein r is t (i', k) represents similarity values of points k before iteration as clustering centers of other points except the i, and all attractions larger than or equal to 0 are taken; r is (r) t (k, k) is the extent to which node k is not suitable to be partitioned into other cluster centers; a, a t+1 (k, k) represents the degree of attribution when i is equal to k.
When updating the attribution degree and the attraction degree, the damping coefficient lambda is introduced for avoiding oscillation, the t+1st time r t+1 (i, k) and a t+1 The iteration value of (i, k) is,
r t+1 (i,k)＝(1-λ)r t+1 (i,k)+λr t (i,k) (4)
a t+1 (i,k)＝(1-λ)a t+1 (i,k)+λa t (i,k) (5)
wherein a is t (i, k) and a t+1 (i, k) represents the values of the t-th and t+1th iterations of a (i, k), respectively; r is (r) t (i, k) and r t+1 (i, k) represents the values of the t-th and t+1th iterations of r (i, k), respectively.
For hyperspectral images with a plurality of wave bands and strong spectrum correlation, the AP clustering algorithm is used for carrying out preliminary clustering on all wave bands of the hyperspectral images, so that the spectrum correlation of an optimal spectrum wave band set selected by an ABS wave band can be effectively reduced, and the reconstructed hyperspectral images contain more main spectrum information in the original hyperspectral images.
in step 102, the adaptive band selection algorithm (Adaptive Band Selection, ABS) takes into account the spatial correlation and spectral correlation of the hyperspectral image, and calculates an inter-band index to select a band combination that has a small correlation and that can maximally contain the main information of the hyperspectral image. The ABS index is calculated by formulas (6), (7) and (8),
in the formulas (6), (7) and (8), sigma i Representing standard deviation of the ith band; i i An ABS index representing the ith band, M and N representing the rows and columns of the hyperspectral image, respectively; f (f) i (x, y) represents an i-th band;
represents the pixel average value of the ith band, R i-1，i And R is i，i+1 The correlation coefficient of the ith wave band and the front wave band and the rear wave band is smaller, which indicates that the higher the independence and the smaller the redundancy between the two wave band data are, and E is a mathematical expectation.
And calculating the ABS indexes of all the wave bands in each cluster, arranging the ABS indexes according to the arrangement from large to small, and selecting the number of the wave bands according to the set threshold or the ranking of the ABS indexes of the wave bands, thereby obtaining the optimal wave band set of the hyperspectral image.
And 104, processing the reconstructed hyperspectral image by utilizing a collaborative representation algorithm to obtain an abnormality detection target.
In step 104, in the collaborative representation algorithm, the to-be-detected pel is approximately represented by using a linear combination of atoms in a background dictionary, the to-be-detected pel is a center pel of the sliding double window, and the background dictionary is composed of pels of the area between the inner window and the outer window. In the assumption ofThe heart element is Z epsilon R n×1 N is the spectral dimension of the hyperspectral image, and the sizes of the outer window and the inner window of the double window are respectively win out And win in Background dictionary X m (size n×m) is composed of pixels between inner and outer windows, where m=win out ×win out -win in ×win in The core of the collaborative representation is to solve for the weight vector beta,
make sum->
At the same time, the minimum is reached, and the objective function of the collaborative representation is as follows:
where η is the Lagrangian multiplier.
Deriving η in the above formula and letting the derivative be equal to 0, the expression for β is:
wherein I is an identity matrix.
In order to adjust the influence of different pixels in the background dictionary on the central pixel, the Tikhonov regularization matrix is utilized to adjust the weight of the pixels in the background dictionary:
reconstructing the pixel to be detected, namely an approximate pixel Z, by using the calculated weight vector beta 1 ＝X m Beta. Then, calculate the pixel Z and its approximation Z 1 Residual r between 1 As shown in formula (12) and compared with a threshold value, if r 1 If the pixel value is larger than the threshold value, the pixel to be detected is an abnormal pixel; if r 1 And if the pixel to be detected is smaller than the threshold value, the pixel to be detected is a background pixel.
r 1 ＝‖Z-Z 1 ‖ 2 (12)
In summary, the algorithm of the invention comprises the following steps:
(a) Input hyperspectral image x= { X 1 ,x 2 ,L,x n ' parameter lambda and inner and outer window size win in And wino ut ；
(b) Clustering all wave bands of the hyperspectral image according to formulas (1) to (5);
(c) Performing ABS index calculation on all wave bands in each cluster according to formulas (6) and (8), and selecting the top n optimal wave bands in all clusters to form a hyperspectral image optimal wave band set;
(d) And (3) performing anomaly detection on the reconstructed hyperspectral image consisting of the hyperspectral image optimal band set according to formulas (9) and (12).
And (3) experimental verification:
to prove the effectiveness of the algorithm, this chapter utilizes two sets of true hyperspectral datasets: the tests Island and Gainesville, and calculate the abnormal detection accuracy of the algorithm under different wave band numbers, and compare with RX, LRX, CRD and LSAD algorithms.
1. Parameter setting
The optimal band number of the algorithm is roughly set to (5,10,15,20,25,30,35,40,45,50), (win out ,win in Lambda) is set to (5,3,10) in the coatings Island dataset -5 ) Set to (7,5,10) in the Gainesville dataset 4 )。
2. Algorithm result comparison and analysis
The anomaly detection AUC values of the algorithm in the coatings Island and Gainesville datasets are shown in FIG. 3, based on the setting of the optimum band number.
According to the abnormal detection AUC value of the algorithm, the abnormal detection accuracy of the algorithm on the Coats Island and Gainesville data sets increases with the increase of the number of the wave bands of the reconstructed hyperspectral image, and starts to slowly decrease after reaching the highest value. The reason is that when the spectrum wave bands are smaller, the reconstructed hyperspectral image has insufficient spectrum information, and the abnormal detection accuracy is lower; when the spectral band is large, the anomaly detection accuracy may be affected by spectral information including noise bands or other low quality bands, resulting in a decrease in the anomaly detection accuracy. For the coatings Island dataset, the anomaly detection accuracy is highest at a spectral band number of 40, auc=0.996; for the Gainesville dataset, the anomaly detection accuracy is highest at a spectral band number of 35, auc=0.980. The AUC values and time of the present and comparative algorithms on both sets of data are shown in table 1.
Table 1 AUC values and time (t/s) for two sets of anomaly detection experiments
As can be seen from the data in table 1, the anomaly detection accuracy of the present algorithm is slightly improved compared with the CRD algorithm, because the present algorithm removes a part of noise bands and other low quality bands in the hyperspectral image, so that the distinguishability of the anomaly information and the background information is enhanced. Meanwhile, compared with the CRD algorithm, the algorithm has the advantages that the operation time is greatly reduced, because the algorithm removes a large amount of redundant information of the hyperspectral image, and the data volume calculated by the algorithm is reduced, so that the operation speed of the algorithm is improved. Compared with other comparison algorithms, the algorithm also keeps higher accuracy and operation speed, and especially compared with an LRX algorithm and an LSAD algorithm, the algorithm consumes less time in operation under the condition that the anomaly detection accuracy is not great. Fig. 4 is a graph showing the comparison of the anomaly detection results for two sets of data sets, and fig. 5 is a graph showing the ROC curve of the anomaly detection results for two sets of data sets.
In another aspect, the present invention further provides a hyperspectral anomaly detection apparatus 200 based on band selection and collaborative representation, including:
a data acquisition module 201, configured to acquire spectral data of all bands of the hyperspectral image;
the clustering module 202 is configured to process spectral data of all bands by using a neighbor propagation clustering algorithm, and obtain a clustering result of the hyperspectral image;
the image reconstruction module 203 is configured to process all bands in each clustering result by using an adaptive band selection algorithm, obtain an optimal band set arranged according to ABS indexes from large to small and the ABS indexes are greater than a preset threshold, and reconstruct a hyperspectral image according to the optimal band set;
the anomaly detection module 204 is configured to process the reconstructed hyperspectral image by using a collaborative representation algorithm to obtain an anomaly detection target.
In yet another embodiment of the present invention, there is also provided an apparatus including a processor and a memory storing at least one instruction, at least one program, a set of codes, or a set of instructions loaded and executed by the processor to implement the distributed data processing cluster high availability method described in the embodiments of the present invention.
In yet another embodiment of the present invention, a computer readable storage medium is provided, where at least one instruction, at least one program, a set of codes, or a set of instructions is stored, where the at least one instruction, the at least one program, the set of codes, or the set of instructions are loaded and executed by a processor to implement the distributed data processing cluster high availability method described in the embodiments of the present invention.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes a plurality of computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of a plurality of available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), etc.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims (4)
1. A hyperspectral anomaly detection method and device are characterized by comprising the following steps:
acquiring spectrum data of all wave bands of a hyperspectral image;
processing the spectrum data of all wave bands by utilizing a neighbor propagation clustering algorithm to obtain a clustering result of the hyperspectral image;
processing all wave bands in each clustering result by utilizing a self-adaptive wave band selection algorithm to obtain an optimal wave band set which is arranged from large to small according to ABS indexes and has the ABS indexes larger than a preset threshold value, and reconstructing the hyperspectral image according to the optimal wave band set;
and processing the reconstructed hyperspectral image by utilizing a collaborative representation algorithm to obtain an anomaly detection target.
2. A hyperspectral anomaly detection device based on band selection and collaborative representation, comprising:
the data acquisition module is used for acquiring spectrum data of all wave bands of the hyperspectral image;
the clustering module is used for processing the spectrum data of all wave bands by utilizing a neighbor propagation clustering algorithm to obtain a clustering result of the hyperspectral image;
the image reconstruction module is used for processing all the wavebands in each clustering result by utilizing a self-adaptive waveband selection algorithm, obtaining an optimal waveband set which is arranged from large to small according to ABS indexes and has the ABS indexes larger than a preset threshold value, and reconstructing the hyperspectral image according to the optimal waveband set;
and the anomaly detection module is used for processing the reconstructed hyperspectral image by utilizing a collaborative representation algorithm to obtain an anomaly detection target.
3. An electronic device comprising a processor and a memory having stored therein at least one instruction, at least one program, code set, or instruction set that is loaded and executed by the processor to implement the band selection and co-representation based hyperspectral anomaly detection method of claim 1.
4. A computer readable storage medium having stored therein at least one instruction, at least one program, code set, or instruction set loaded and executed by a processor to implement the band selection and synergistic presentation based hyperspectral anomaly detection method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310324310.8A CN116416231A (en) | 2023-03-29 | 2023-03-29 | Hyperspectral anomaly detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310324310.8A CN116416231A (en) | 2023-03-29 | 2023-03-29 | Hyperspectral anomaly detection method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116416231A true CN116416231A (en) | 2023-07-11 |
Family
ID=87059060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310324310.8A Pending CN116416231A (en) | 2023-03-29 | 2023-03-29 | Hyperspectral anomaly detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116416231A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117435940A (en) * | 2023-12-20 | 2024-01-23 | 龙建路桥股份有限公司 | Spectrum detection method for winter concrete curing process |
-
2023
- 2023-03-29 CN CN202310324310.8A patent/CN116416231A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117435940A (en) * | 2023-12-20 | 2024-01-23 | 龙建路桥股份有限公司 | Spectrum detection method for winter concrete curing process |
| CN117435940B (en) * | 2023-12-20 | 2024-03-05 | 龙建路桥股份有限公司 | Spectrum detection method for winter concrete curing process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wyatt et al. | Anoddpm: Anomaly detection with denoising diffusion probabilistic models using simplex noise | |
| CN108198147B (en) | Multi-source image fusion denoising method based on discriminant dictionary learning | |
| CN106097278B (en) | Sparse model, reconstruction method and dictionary training method of multi-dimensional signal | |
| EP2245577A1 (en) | Parallel and adaptive signal processing | |
| CN110648290A (en) | Dual-core non-local mean image denoising method based on SURE parameter optimization | |
| CN116416231A (en) | Hyperspectral anomaly detection method and device | |
| CN111507047B (en) | Inverse scattering imaging method based on SP-CUnet | |
| CN114787826A (en) | Generating high-dimensional high-utility synthetic data | |
| CN110415180B (en) | SAR image denoising method based on wavelet convolution neural network | |
| CN116402825A (en) | Bearing fault infrared diagnosis method, system, electronic equipment and storage medium | |
| CN110910325A (en) | Medical image processing method and device based on artificial butterfly optimization algorithm | |
| Cammarasana et al. | A universal deep learning framework for real-time denoising of ultrasound images | |
| CN113327205A (en) | Phase denoising network and method based on convolutional neural network | |
| CN115330901B (en) | Image reconstruction method and device based on compressed sensing network | |
| CN110929731B (en) | Medical image processing method and device based on pathfinder intelligent search algorithm | |
| CN112782115A (en) | Method for detecting consistency of sensory characteristics of cigarettes based on near infrared spectrum | |
| CN109460772B (en) | Spectral band selection method based on information entropy and improved determinant point process | |
| CN116543259A (en) | Deep classification network noise label modeling and correcting method, system and storage medium | |
| CN112951320B (en) | Biomedical network association prediction method based on ensemble learning | |
| Li et al. | Low-dose ct image denoising based on improved WGAN-GP | |
| CN113177608A (en) | Neighbor model feature selection method and device for incomplete data | |
| CN111797732A (en) | Video motion identification anti-attack method insensitive to sampling | |
| Park et al. | Colonoscopy video quality assessment using hidden Markov random fields | |
| CN117095208B (en) | Lightweight scene classification method for photoelectric pod reconnaissance image | |
| Cheng et al. | MCDIP‐ADMM: Overcoming overfitting in DIP‐based CT reconstruction |
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 |