Burn scar detection using cloudy MODIS images via low-rank and sparsity-based models
The DOI matches its Crossref record; 20 of the 20 references it cites were found in a registry or an index.
Abstract
In this paper, we propose a general framework to detect burned area using multiple partially occluded Moderate Resolution Imaging Spectroradiometer (MODIS) images. By treating each MODIS image as the superimposition of 3 layers: background, burned area, and cloud, we first apply a low-rank and sparse matrix decomposition technique known as Robust Principal Component Analysis (RPCA) to separate cloud from the other two components. Secondly, a joint sparsity-based representation for classification algorithm is employed to determine the burned area. We further robustify the algorithm by proposing a novel combined model that can separate and suppress the cloud information while simultaneously detecting the burn scar. Preliminary results using actual MODIS images clearly demonstrated the efficacy of the proposed technique. The probability of correct detection of burned area can reach more than 80% with less than 10% false alarm rate, even in an extremely cloudy day.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“It is worth noting that HSIs are only a mixture of several endmembers due to the limitation of the category numbers of materials in real reality [10].”
From Background Reconstruction via 3D-Transformer Network for Hyperspectral Anomaly Detection (2023) · cites this paper as [10]
Citations
Cites
Donoho, D. L. (2006) · IEEE Transactions on Information Theory · doi:10.1109/tit.2006.871582 · Cited by 23,994
Boyd, S., Parikh, N., Chu, E., Peleato, B., & Eckstein, J. (2011) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000016 · Cited by 17,075
Boyd, S. (2010) · now publishers, Inc. eBooks · doi:10.1561/9781601984616 · Cited by 12,677
Robust principal component analysis?
Candès, E. J., Li, X., Ma, Y., & Wright, J. (2011) · Journal of the ACM · doi:10.1145/1970392.1970395 · Cited by 6,921 · cited 3×
“The sparse coefficients of these measurements can be recovered jointly by solving the following 12-regularized minimization [10]: min S ‖S‖12 s.t.Y =DS, (2) where the norm ‖S‖12 is defined as the sum of the 2-norm of rows of S .”
BackgroundMethodA Singular Value Thresholding Algorithm for Matrix Completion
Cai, J.-F., Candès, E. J., & Shen, Z. (2010) · SIAM Journal on Optimization · doi:10.1137/080738970 · Cited by 5,809 · cited 1×
“After suppressing the cloud information in every frame, the next step is to automatically classify every pixel if it belongs to a burn scar or non-burn-scar area using a sparsity-based representation model.”
MethodClassification of hyperspectral remote sensing images with support vector machines
Melgani, F., & Bruzzone, L. (2004) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2004.831865 · Cited by 4,411 · cited 2×
“Detection performance is validated on a real MODIS imaging dataset; which clearly demonstrates the efficacy of the proposed technique compared to conventional detection algorithms such as Adaptive Matched Subspace Detector (AMSD) [7] and Support Vector Machine (SVM) [8].”
BackgroundMethodFigueiredo, M. A. T., Nowak, R. D., & Wright, S. J. (2007) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2007.910281 · Cited by 3,524
Wright, J. N., Ganesh, A., Rao, S., Peng, Y., & Ma, Y. (2009) · Illinois Digital Environment for Access to Learning and Scholarship (University of Illinois at Urbana-Champaign) · Cited by 1,495
An active-fire based burned area mapping algorithm for the MODIS sensor
Giglio, L., Loboda, T., Roy, D. P., Quayle, B., & Justice, C. O. (2009) · Remote Sensing of Environment · doi:10.1016/j.rse.2008.10.006 · Cited by 657
A Fast Algorithm for Edge-Preserving Variational Multichannel Image Restoration
Yang, J., Yin, W., Zhang, Y., & Wang, Y. (2009) · SIAM Journal on Imaging Sciences · doi:10.1137/080730421 · Cited by 529
Research works printed but not linked: 2
- li, 0, Identify anomaly component by sparsity and low rank, IEEE International Conf on Image Processing (ICIP)
- 0, Applied research 11c (arllc) firemap data product
Cited by
Deep Learning and Transformer Approaches for UAV-Based Wildfire Detection and Segmentation
Ghali, R., Akhloufi, M. A., & Mseddi, W. S. (2022) · Sensors · doi:10.3390/s22051977 · Cited by 172
Wildfire and Smoke Detection Using Staged YOLO Model and Ensemble CNN
Bahhar, C., Ksibi, A., Ayadi, M., Jamjoom, M. M., Ullah, Z., Soufiene, B. O., & Sakli, H. (2023) · Electronics · doi:10.3390/electronics12010228 · Cited by 99
“The authors developed a 360-degree fire detection dataset comprising 150 equirectangular photos.”
Hyperspectral Anomaly Detection With Tensor Average Rank and Piecewise Smoothness Constraints
Sun, S., Liu, J., Chen, X., Li, W., & Li, H. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3152252 · Cited by 84
“There are various ways to construct a dictionary [60], [61].”
Forest Fire Identification in UAV Imagery Using X-MobileNet
Namburu, A., Selvaraj, P., Mohan, S., Ragavanantham, S., & Eldin, E. T. (2023) · Electronics · doi:10.3390/electronics12030733 · Cited by 68
Discriminative Reconstruction for Hyperspectral Anomaly Detection With Spectral Learning
Lei, J., Fang, S., Xie, W., Li, Y., & Chang, C. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.2982406 · Cited by 68
“The background joint sparse representation detector (BJSRD) [35] measures the length of the matched projection of the tested pixel on the orthogonal complementary background subspace by using the joint sparse representation (JSR) [36] to detect anomalous pixels.”
MethodLin, J., Huang, T., Zhao, X.-L., Chen, Y., Zhang, Q., & Yuan, Q. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3140800 · Cited by 63
“However, in the real scenario, all spectral bands of RS images are usually contaminated by large-scale thick clouds, which means that only using spatial or spectral information is not enough to reconstruct underlying information [26], [27].”
BackgroundDeep Learning for Land Cover Classification Using Only a Few Bands
Kwan, C., Ayhan, B., Budavari, B., Lu, Y., Pérez, D., Li, J., Bernabé, S., & Plaza, A. (2020) · Remote Sensing · doi:10.3390/rs12122000 · Cited by 63
“Hyperspectral data have been used for chemical agent detection and classification [1,2], small target detection [3,4], fire damage assessment [5,6], anomaly detection [7–13], border monitoring [14], change detection [15–18], and mineral map abundance estimation on Mars [19,20].”
MethodSelf-citationZhu, L., & Wen, G. (2018) · Remote Sensing · doi:10.3390/rs10020272 · Cited by 58
“The background joint sparse representation detector (BJSRD) [24] measures the length of the matched projection on the orthogonal complementary background subspace for the pixel under test by joint sparse representation (JSR) [25,26].”
MethodKwan, C. (2019) · Information · doi:10.3390/info10110353 · Cited by 58
“MS and HS images have been widely used in fire damage assessment [5], anomaly detection [6–10], chemical agent detection and classification [11], border monitoring [12], target detection [13–15], and change detection [16,17].”
BackgroundMethodSelf-citationCheng, T., & Wang, B. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.3004478 · Cited by 58
“In sparse representation for supervised target detection, the union dictionary consisting of the background dictionary and target dictionary is usually given a priori [32], [33].”
BackgroundA joint sparsity approach to tunnel activity monitoring using high resolution satellite images
Dao, M., Kwan, C., Koperski, K., & Marchisio, G. (2017) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2017.8249061 · Cited by 57
“5 visualizes the soil target detection model via joint sparse representation (JSR) [15][16].”
BackgroundSelf-citationA Novel Deep Fully Convolutional Network for PolSAR Image Classification
Li, Y., Chen, Y., Liu, G., & Jiao, L. (2018) · Remote Sensing · doi:10.3390/rs10121984 · Cited by 46
Deep learning for effective detection of excavated soil related to illegal tunnel activities
Pérez, D., Banerjee, D., Kwan, C., Dao, M., Shen, Y., Koperski, K., Marchisio, G., & Jiang, L. (2017) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2017.8249062 · Cited by 40
“Motivated by a joint sparsity approach [7], [9], [10], which enforces group sparsity across neighboring pixels, we also applied a morphological closing operation filter to the probability map.”
MethodSelf-citationMulti-temporal cloud detection based on robust PCA for optical remote sensing imagery
Zhang, H., Huang, Q., Zhai, H., & Zhang, L. (2021) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106342 · Cited by 39
Kwan, C., Gribben, D., Ayhan, B., Bernabé, S., Plaza, A., & Selva, M. (2020) · Remote Sensing · doi:10.3390/rs12091392 · Cited by 36
Self-citationGroup Low-Rank Nonnegative Matrix Factorization With Semantic Regularizer for Hyperspectral Unmixing
Wang, M., Zhang, B., Pan, X., & Yang, S. (2018) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2018.2805779 · Cited by 31
“Very recently, the low-rank and sparse property of hyperspectral data are also explored [23], [24], including low-rank and sparse prior [34], high-order lowrank prior [35] and spatial prior [36], [37], which are able to find more compact representation of hyperspectral data.”
BackgroundFusion of themis and TES for accurate Mars surface characterization
Kwan, C., Ayhan, B., & Budavari, B. (2017) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss.2017.8127723 · Cited by 30
Self-citationCompressive Vehicle Tracking Using Deep Learning
Kwan, C., Chou, B., Echavarren, A., Budavari, B., Jiang, L., & Tran, T. D. (2018) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2018.8796778 · Cited by 25
Self-citationNew sparsity based pansharpening algorithms for hyperspectral images
Kwan, C., Budavari, B., Dao, M., & Zhou, J. (2017) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2017.8248993 · Cited by 23
“Other notable applications of multispectral and hyperspectral images include target detection [6]-[9], anomaly and change detection [10]-[17], tunnel monitoring [18]-[19], and Mars exploration [20][21].”
BackgroundSelf-citationClelland, A. A., Marshall, G. J., Baxter, R., Potter, S., Talucci, A. C., Rady, J. M., Genet, H., Rogers, B. M., & Natali, S. M. (2024) · Remote Sensing · doi:10.3390/rs16173306 · Cited by 23
“All, however, have to cope with the presence of clouds, which can complicate the process by obscuring the land for days, if not weeks, at a time and can be easily confused with smoke from fires [53,54].”
BackgroundRemote Sensing Performance Enhancement in Hyperspectral Images
Kwan, C. (2018) · Sensors · doi:10.3390/s18113598 · Cited by 22
“Remote sensing using multispectral (MS) and hyperspectral (HS) images can help fire detection [1], anomaly detection [2–6], chemical agent detection [7], border monitoring [8], target detection [9–11], and change detection [12–14].”
BackgroundSelf-citationCombining Satellite Images with Feature Indices for Improved Change Detection
Pérez, D., Lu, Y., Kwan, C., Shen, Y., Koperski, K., & Jiang, L. (2018) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2018.8796538 · Cited by 19
“The goal of the post-processing filters is to enforce group sparsity across neighboring pixels, which has significantly improved the results in similar works [3], [17].”
BackgroundResultSelf-citationWildfire Detection Based on the Spatiotemporal and Spectral Features of Himawari-8 Data
Zheng, Z., Hu, H., Huang, W., Zhou, F., Ma, Y., Liu, Q., Jiang, L., & Wang, S. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3434434 · Cited by 19
“Imaging sensors on Sun-synchronous satellites, such as the Moderate Resolution Imaging Spectroradiometer (MODIS), Advanced Very High Resolution Radiometer (AVHRR), and Visible Infrared Imaging Radiometer Suite (VIIRS) [20], [21], [22], [23], offer high spatial resolution but low temporal resolution.”
BackgroundA new nonlinear change detection approach based on band ratioing
Kwan, C., Ayhan, B., & Zhou, J. (2018) · Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery · doi:10.1117/12.2303648 · Cited by 18
Self-citationChen, S.-Y., Lin, C., Li, G.-J., Hsu, Y., & Liu, K.-H. (2021) · Sensors · doi:10.3390/s21062077 · Cited by 14
“More researches about RPCA or sparse representation can be found in References [71–75].”
BackgroundEvaluation of Combining Optical and SAR Imagery for Burned Area Mapping using Machine Learning
Lestari, A. I., Rizkinia, M., & Sudiana, D. (2021) · IEEE Annual Computing and Communication Workshop and Conference (CCWC) · doi:10.1109/ccwc51732.2021.9376117 · Cited by 14
“An approach has been conducted to obtain burned scar detection in a cloudy, low spatial resolution image using joint sparse representation and low rank simultaneously [24].”
MethodBackground Reconstruction via 3D-Transformer Network for Hyperspectral Anomaly Detection
Wu, Z., & Wang, B. (2023) · Remote Sensing · doi:10.3390/rs15184592 · Cited by 13
High performance image completion using sparsity based algorithms
Kwan, C., & Zhou, J. (2018) · SPIE Computational Imaging · doi:10.1117/12.2303661 · Cited by 12
Self-citationA Joint Sparsity Approach to Soil Detection Using Expanded Bands of WV-2 Images
Dao, M., Kwan, C., Bernabé, S., Plaza, A., & Koperski, K. (2019) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2019.2911923 · Cited by 12
“For example, for data collected on December 2, 2010, the JSR detection accuracy with EMAP is 87.06%, which is 24% more than that obtained using the original 8-band data.”
BackgroundInfluentialSelf-citationImage Resolution Enhancement for Remote Sensing Applications
Kwan, C. (2018) · International Conference on Vision, Image and Signal Processing (ICVISP) · doi:10.1145/3271553.3271590 · Cited by 12
“After a careful study of the fusion problem between MODIS and Landsat, it turns out that pansharpening algorithms, such as the algorithms mentioned in [6,29], cannot be applied.”
BackgroundMethodInfluentialSelf-citationPerceptually Lossless Compression with Error Concealment for Periscope and Sonar Videos
Kwan, C., Larkin, J., Budavari, B., Shang, E. L., & Tran, T. D. (2019) · Signal & Image Processing An International Journal · doi:10.5121/sipij.2019.10201 · Cited by 11
Self-citationXu, S., Wang, J., & Wang, J. (2023) · Remote Sensing · doi:10.3390/rs16010152 · Cited by 11
“Similarly, Dao et al. used a similar idea to model a cloud as sparse noise and successfully applied the burn scar detection algorithms for cloudy images [27].”
Chen, S.-Y., Lin, C., Chuang, S.-J., & Kao, Z.-Y. (2019) · Remote Sensing · doi:10.3390/rs11091081 · Cited by 11
Tracking of Multiple Pixel Targets Using Multiple Cameras
Zhou, J., & Kwan, C. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-92537-0_56 · Cited by 10
Self-citationMissing Link Prediction in Social Networks
Zhou, J., & Kwan, C. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-92537-0_40 · Cited by 10
Self-citationChange detection using Landsat and Worldview images
Kwan, C., Chou, B., Hagen, L., Pérez, D., Shen, Y., Li, J., & Koperski, K. (2019) · Algorithms, Technologies, and Applications for Multispectral and Hyperspectral Imagery · doi:10.1117/12.2518515 · Cited by 9
“The RX detector is given by 1 ( ) T D x x R x (3) where R is the covariance of the background.”
MethodSelf-citationWu, Z., & Wang, B. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3218826 · Cited by 9
“In order to express other anomalies in the hyperspectral imagery, the anomaly dictionary is composed of anomalies with significantly different spectra from their surroundings [17], [18], [31].”
BackgroundMethodInfluentialPerceptually Lossless Video Compression with Error Concealment
Kwan, C., Shang, E. L., & Tran, T. D. (2018) · International Conference on Vision, Image and Signal Processing (ICVISP) · doi:10.1145/3271553.3271622 · Cited by 8
Self-citationShi, J., Yang, S., Cui, S., Luo, T., Li, X., Lu, W., & Han, L. (2023) · Earth and Space Science · doi:10.1029/2022ea002738 · Cited by 7
“RPCA is an unsupervised subspace learning technique that focuses on accurately recovering low-rank subspaces from sparse noise (Dao et al., 2016).”
BackgroundHuo, Y., Zhang, Y., Xu, J., Dai, X., Shen, L., Liu, C., & Fang, X. (2025) · Energies · doi:10.3390/en18061511 · Cited by 6
Ground object detection in worldview images
Kwan, C., Chou, B., Gribben, D., Hagen, L., Yang, J. M., Ayhan, B., & Koperski, K. (2019) · Signal processing, sensor fusion, and target recognition. Conference · doi:10.1117/12.2518529 · Cited by 3
Self-citationA New Pansharpening Approach for Hyperspectral Images
Kwan, C., Zhou, J., & Budavari, B. (2018) · InTech eBooks · doi:10.5772/intechopen.71023 · Cited by 2
Self-citationDeep learning approach for uav-based wildfire detection and segmentation
Sreenath, K., Hemalatha Reddy, G., Venkata Ranga Reddy, K., Sneha Sandhya, G., & D, M. B. M. (2022) · South Asian Journal of Engineering and Technology · doi:10.26524/sajet.2022.12.45 · Cited by 1
On Sampled Sequence Representations at Discontinuities and Their Impact on Discrete Convolution
Kwan, C. (2026) · Electronics · doi:10.3390/electronics15132962
“…Fourier transform (FFT) in the 1960s [1] and the discrete cosine transform (DCT) in the 1970s [2], digital and image processing algorithms have been widely used in many applications, including acoustic processing [3], voice recognition [4], face recognition [5], remote sensing [6], and many others.”
Self-citation
Metadata
- Authors
- Dao, M., Kwan, C., Ayhan, B., & Tran, T. D.
- Affiliations
- Johns Hopkins University (US)
- Applied Research (United States) (US)
- Venue
- 2016 IEEE Global Conference on Signal and Information Processing (GlobalSIP)
- Publisher
- IEEE
- Published
- 2016-12
- Pages
- 177-181
- ISSN
- 2376-4066 (ISSN-L)
- OpenAlex
- W2610484101
- Cited by
- 42
- References
- 20
- Citations by year
- 2026: 1 · 2025: 1 · 2024: 2 · 2023: 5 · 2022: 4 · 2021: 3 · 2020: 4 · 2019: 6 · 2018: 12 · 2017: 4
- FWCI
- 12.02
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Global Conference on Signal and Information Processing (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Remote-Sensing Image Classification
- Physical Sciences › Engineering › Safety, Risk, Reliability and Quality › Fire Detection and Safety Systems
- Physical Sciences › Environmental Science › Global and Planetary Change › Fire effects on ecosystems
- Keywords
- burned area detection, MODIS, cloud removal, robust principal component analysis, low-rank and sparse decomposition, joint sparsity, cloud occlusion, fire detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.