Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8900496
B. Scarino, D. Doelling, C. Haney, R. Bhatt, A. Gopalan
Earth-viewed images acquired during a recent asteroid-intercept mission present a unique opportunity for radiometric calibration of visible imagers onboard a space exploration probe. Measurements from the CERES-consistent DSCOVR-EPIC imager act as a reference in providing spatially, temporally, and angularly matched radiance values for deriving OSIRIS-REx-NavCam sensor calibration gains. The calibration is accomplished using an optimized all-sky tropical ocean ray-matching technique, which employs complex pixel remapping, navigation correction, and angular geometry consideration. Of critical consideration in this specific inter-calibration event is the extreme difference in spectral response function (SRF) width between the NavCam and EPIC imagers, which could cause a rather large bias. The NASA-LaRC SCIAMACHY-based online spectral band adjustment factor (SBAF) calculation tool provides an empirical solution to such potential spectral-difference-induced biases through a high-spectral-resolution hyperspectral convolution approach. The adjustments produced from this tool can effectively reduce the calibration gain bias of NavCam2 by nearly 6%, thereby adjusting the NavCam2 sensor to within 3.2% of its pre-launch calibration. These results highlight the capability of the SBAF tool to account for exceptionally disparate SRFs.
{"title":"Extreme Case of Spectral Band Difference Correction Between the Osiris-Rex-Navcam2 Dscovr-Epic Imagers","authors":"B. Scarino, D. Doelling, C. Haney, R. Bhatt, A. Gopalan","doi":"10.1109/IGARSS.2019.8900496","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8900496","url":null,"abstract":"Earth-viewed images acquired during a recent asteroid-intercept mission present a unique opportunity for radiometric calibration of visible imagers onboard a space exploration probe. Measurements from the CERES-consistent DSCOVR-EPIC imager act as a reference in providing spatially, temporally, and angularly matched radiance values for deriving OSIRIS-REx-NavCam sensor calibration gains. The calibration is accomplished using an optimized all-sky tropical ocean ray-matching technique, which employs complex pixel remapping, navigation correction, and angular geometry consideration. Of critical consideration in this specific inter-calibration event is the extreme difference in spectral response function (SRF) width between the NavCam and EPIC imagers, which could cause a rather large bias. The NASA-LaRC SCIAMACHY-based online spectral band adjustment factor (SBAF) calculation tool provides an empirical solution to such potential spectral-difference-induced biases through a high-spectral-resolution hyperspectral convolution approach. The adjustments produced from this tool can effectively reduce the calibration gain bias of NavCam2 by nearly 6%, thereby adjusting the NavCam2 sensor to within 3.2% of its pre-launch calibration. These results highlight the capability of the SBAF tool to account for exceptionally disparate SRFs.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"10 1","pages":"8996-8998"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77764677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8900078
M. Younis, F. Almeida, S. Huber, M. Zonno, M. Rodríguez-Cassola, S. Hensley, G. Krieger
Utilizing digital multi-channel technology, spacebome synthetic aperture radar instruments are capable of imaging swath widths of hundreds of kilometers at fine azimuth resolution. The main benefit follows through the extension of the trade space and the use of new digital beam-forming techniques facilitated through the multi-channel instrument architecture. This is truly a quantum leap as the performance of these systems will be orders of magnitude better than current in-orbit and state-of-the art systems. One of the basic restrictions applicable to spaceborne platforms hosting both the transmitter and receiver is the "blinding" of the receiver during the transmit time instances, which manifests itself through imaging gaps. One of the main challenges the instrument designers are faced with, is to circumvent these gaps, requiring the use of dedicated instrument operation modes. An alternative approach is multi-beam imaging, i.e. to allow the gaps in the single SAR acquisition, while using an appropriate mission design for filling the blind gaps. This paper explores the trade space options for high-resolution wide-swath SAR imaging. The comparison of multi-beam and gapless imaging from an instrument design and performance point of view is elaborated.
{"title":"The Cost of Opportunity for Gapless Imaging","authors":"M. Younis, F. Almeida, S. Huber, M. Zonno, M. Rodríguez-Cassola, S. Hensley, G. Krieger","doi":"10.1109/IGARSS.2019.8900078","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8900078","url":null,"abstract":"Utilizing digital multi-channel technology, spacebome synthetic aperture radar instruments are capable of imaging swath widths of hundreds of kilometers at fine azimuth resolution. The main benefit follows through the extension of the trade space and the use of new digital beam-forming techniques facilitated through the multi-channel instrument architecture. This is truly a quantum leap as the performance of these systems will be orders of magnitude better than current in-orbit and state-of-the art systems. One of the basic restrictions applicable to spaceborne platforms hosting both the transmitter and receiver is the \"blinding\" of the receiver during the transmit time instances, which manifests itself through imaging gaps. One of the main challenges the instrument designers are faced with, is to circumvent these gaps, requiring the use of dedicated instrument operation modes. An alternative approach is multi-beam imaging, i.e. to allow the gaps in the single SAR acquisition, while using an appropriate mission design for filling the blind gaps. This paper explores the trade space options for high-resolution wide-swath SAR imaging. The comparison of multi-beam and gapless imaging from an instrument design and performance point of view is elaborated.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"98 1","pages":"8316-8319"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76612018","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8900558
Gabriel Vasile
The Independent Component Analysis (ICA) has been recently introduced as a reliable alternative to identify canonical scattering mechanisms within PolSAR images. This paper addresses an important aspect for applying such methods on real data, namely statistical classification with ICA. A novel algorithm is proposed by adjusting the iterative segmentation from [1], [2] to the particular nature of the Touzi’s polarimetric decomposition [3]. This algorithm is tested using P-band airborne PolSAR data acquired for the ESA campaign TropiSAR campaign.
{"title":"On ICA Based ICTD Classification of Polsar Data","authors":"Gabriel Vasile","doi":"10.1109/IGARSS.2019.8900558","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8900558","url":null,"abstract":"The Independent Component Analysis (ICA) has been recently introduced as a reliable alternative to identify canonical scattering mechanisms within PolSAR images. This paper addresses an important aspect for applying such methods on real data, namely statistical classification with ICA. A novel algorithm is proposed by adjusting the iterative segmentation from [1], [2] to the particular nature of the Touzi’s polarimetric decomposition [3]. This algorithm is tested using P-band airborne PolSAR data acquired for the ESA campaign TropiSAR campaign.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"8 1","pages":"5129-5132"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81796212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8898430
Y. Deville, Audrey Minghelli, X. Briottet, V. Serfaty, S. Brezini, Fatima Zohra Benhalouche, M. S. Karoui, M. Guillaume, X. Lenot, B. Lafrance, M. Chami, S. Jay
In a very recent paper, we introduced (i) a specific hyper-spectral mixing model for the sea bottom, based on a detailed physical analysis which includes the adjacency effect, and (ii) an associated unmixing method, which is not blind in the sense that it requires a prior estimation of various parameters of that mixing model. We here proceed much further, by first analytically showing that this model can be seen as a specific member of the general class of mixing models involving spectral variability. Therefore, we then process such data with the IP-NMF and UP-NMF blind unmixing methods that we recently proposed in other works to handle spectral variability. Such a variability especially occurs when sea depth significantly varies over the considered scene, and we show that IP-NMF and UP-NMF then yield significantly better pure spectra estimation than a classical method from the literature which was not designed to handle such a variability.
{"title":"Hyperspectral Oceanic Remote Sensing With Adjacency Effects: From Spectral-Variability-Based Modeling To Performance Of Associated Blind Unmixing Methods","authors":"Y. Deville, Audrey Minghelli, X. Briottet, V. Serfaty, S. Brezini, Fatima Zohra Benhalouche, M. S. Karoui, M. Guillaume, X. Lenot, B. Lafrance, M. Chami, S. Jay","doi":"10.1109/IGARSS.2019.8898430","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8898430","url":null,"abstract":"In a very recent paper, we introduced (i) a specific hyper-spectral mixing model for the sea bottom, based on a detailed physical analysis which includes the adjacency effect, and (ii) an associated unmixing method, which is not blind in the sense that it requires a prior estimation of various parameters of that mixing model. We here proceed much further, by first analytically showing that this model can be seen as a specific member of the general class of mixing models involving spectral variability. Therefore, we then process such data with the IP-NMF and UP-NMF blind unmixing methods that we recently proposed in other works to handle spectral variability. Such a variability especially occurs when sea depth significantly varies over the considered scene, and we show that IP-NMF and UP-NMF then yield significantly better pure spectra estimation than a classical method from the literature which was not designed to handle such a variability.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"2014 1","pages":"282-285"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87760527","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8898346
Mohamad Jouni, M. Mura, P. Comon
Image classification has been at the core of remote sensing applications. Optical remote sensing imaging systems naturally acquire images with spectral features corresponding to pixels. Spectral classification ignores the spatial distribution of the data which is becoming more relevant with the development of spatial resolution sensors, and many works aim to incorporate spatial features based on neighborhood through for example, Mathematical Morphology (MM). Additionally, one could stack multiple morphological transformations of the image resulting in a highly complex block of data. Since classification is a tool that requires a matrix of samples and features, and simply stacking the different sets of features can lead to the problem of high dimensionality, we propose a way to create a matrix of low dimensional feature space by modeling the data as tensors and thanks to Canonical Polyadic (CP) decomposition. Experiments on real image show the effectiveness of the proposed method.
{"title":"Hyperspectral Image Classification Using Tensor CP Decomposition","authors":"Mohamad Jouni, M. Mura, P. Comon","doi":"10.1109/IGARSS.2019.8898346","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8898346","url":null,"abstract":"Image classification has been at the core of remote sensing applications. Optical remote sensing imaging systems naturally acquire images with spectral features corresponding to pixels. Spectral classification ignores the spatial distribution of the data which is becoming more relevant with the development of spatial resolution sensors, and many works aim to incorporate spatial features based on neighborhood through for example, Mathematical Morphology (MM). Additionally, one could stack multiple morphological transformations of the image resulting in a highly complex block of data. Since classification is a tool that requires a matrix of samples and features, and simply stacking the different sets of features can lead to the problem of high dimensionality, we propose a way to create a matrix of low dimensional feature space by modeling the data as tensors and thanks to Canonical Polyadic (CP) decomposition. Experiments on real image show the effectiveness of the proposed method.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"51 1","pages":"1164-1167"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81724727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8899275
A. Olioso, X. Briottet, S. Fabre, F. Jacob, A. Michel, S. Nativel, V. Rivalland, J. Roujean
Land surface emissivity is required for deriving surface temperature from thermal infrared radiances. When using single-channel or two-channel thermal infrared sensors, information on emissivity may be derived from spectral reflectance measurements through regression models. In this study, we present relationships derived over bare soils for Landsat 7 – ETM+ sensor. Reflectances in ETM+ channels were obtained from soil spectra (between 0.4 and 13 μm) extracted from the ASTER spectral library and the dataset acquired by Lesaignoux et al. (2013). The best relations were obtained between reflectances in the mid-infrared channels (ETM5 and ETM7) and the thermal infrared channel (ETM6) with correlation coefficients of 0.63 and 0.72 respectively. The relations were mostly generated by the variations of soil reflectances due to changes in soil moisture. Correlations were lower when considering the variations due to soil type.
地表发射率是根据热红外辐射推算地表温度的必要条件。当使用单通道或双通道热红外传感器时，可以通过回归模型从光谱反射率测量中获得发射率信息。在本研究中，我们展示了Landsat 7 - ETM+传感器在裸露土壤上的关系。ETM+通道的反射率由ASTER光谱库和Lesaignoux et al.(2013)获取的数据集提取的土壤光谱(0.4 ~ 13 μm)获得。中红外通道(ETM5和ETM7)与热红外通道(ETM6)的反射率关系最佳，相关系数分别为0.63和0.72。这种关系主要是由土壤水分变化引起的土壤反射率变化引起的。考虑土壤类型差异时，相关性较低。
{"title":"Relations Between Landsat Spectral Reflectances and Land Surface Emissivity Over Bare Soils","authors":"A. Olioso, X. Briottet, S. Fabre, F. Jacob, A. Michel, S. Nativel, V. Rivalland, J. Roujean","doi":"10.1109/IGARSS.2019.8899275","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8899275","url":null,"abstract":"Land surface emissivity is required for deriving surface temperature from thermal infrared radiances. When using single-channel or two-channel thermal infrared sensors, information on emissivity may be derived from spectral reflectance measurements through regression models. In this study, we present relationships derived over bare soils for Landsat 7 – ETM+ sensor. Reflectances in ETM+ channels were obtained from soil spectra (between 0.4 and 13 μm) extracted from the ASTER spectral library and the dataset acquired by Lesaignoux et al. (2013). The best relations were obtained between reflectances in the mid-infrared channels (ETM5 and ETM7) and the thermal infrared channel (ETM6) with correlation coefficients of 0.63 and 0.72 respectively. The relations were mostly generated by the variations of soil reflectances due to changes in soil moisture. Correlations were lower when considering the variations due to soil type.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"6 1","pages":"6937-6940"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81812520","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8898398
R. Reichle, Qing Liu, G. Lannoy, W. Crow, L. Jones, J. Kimball, R. Koster
Global, 3-hourly, 9-km resolution soil moisture estimates are available with a mean latency of ~2.5 days from the NASA Soil Moisture Active Passive (SMAP) mission Level-4 Soil Moisture (L4_SM) product. These estimates are based on the assimilation of SMAP radiometer brightness temperature (Tb) observations into the NASA Catchment land surface model using a spatially distributed ensemble Kalman filter. Routine monitoring of the L4_SM system’s assimilation diagnostics revealed occasionally large observation-minus-forecast Tb differences across eastern central Australia that resulted in large analysis increments (or adjustments) of the model forecast soil moisture. Because this region lacks in situ soil moisture measurements, we developed an alternative approach to assess the veracity of the soil moisture analysis increments in the L4_SM system. Using regional gauge-based precipitation data, we demonstrate that the L4_SM soil moisture increments are correlated with errors in the L4_SM precipitation forcing, suggesting that the SMAP Tb observations contribute valuable information to the L4_SM soil moisture estimates.
{"title":"Verification of the SMAP Level-4 Soil Moisture Analysis Using Rainfall Observations in Australia","authors":"R. Reichle, Qing Liu, G. Lannoy, W. Crow, L. Jones, J. Kimball, R. Koster","doi":"10.1109/IGARSS.2019.8898398","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8898398","url":null,"abstract":"Global, 3-hourly, 9-km resolution soil moisture estimates are available with a mean latency of ~2.5 days from the NASA Soil Moisture Active Passive (SMAP) mission Level-4 Soil Moisture (L4_SM) product. These estimates are based on the assimilation of SMAP radiometer brightness temperature (Tb) observations into the NASA Catchment land surface model using a spatially distributed ensemble Kalman filter. Routine monitoring of the L4_SM system’s assimilation diagnostics revealed occasionally large observation-minus-forecast Tb differences across eastern central Australia that resulted in large analysis increments (or adjustments) of the model forecast soil moisture. Because this region lacks in situ soil moisture measurements, we developed an alternative approach to assess the veracity of the soil moisture analysis increments in the L4_SM system. Using regional gauge-based precipitation data, we demonstrate that the L4_SM soil moisture increments are correlated with errors in the L4_SM precipitation forcing, suggesting that the SMAP Tb observations contribute valuable information to the L4_SM soil moisture estimates.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"12 1","pages":"5363-5366"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87455367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8900289
Xing Peng, Yue Huang, L. Ferro-Famil, Jianjun Zhu, Yanan Du, Haiqiang Fu
Polarimetric synthetic aperture radar tomography (Pol-TomoSAR) allows to achieve a 3-D characterization over urban areas using multiple polarimetric acquisitions. However, using spaceborne datasets, such as TerraSAR-X, it is difficult to localize the distributed or uncorrelated scattering patterns along elevation due to the temporal decorrelation. In order to overcome this limitation, this paper proposes polarimetric correlation tomographic techniques based on Tandem-mode images. The key of this technique is to build a covariance matrix from the observed Tandem coherence pairs, and then apply conventional covariance-based tomographic techniques. This processing allows to extract both coherent and distributed scatterers. The resulting 3-D reconstruction is more refined and detailed, compared to the one derived from TerraSAR-X data. Seven TSX/TDX pairs in fully polarimetric mode over a small county in Yunnan province, China, are used to demonstrate the effectiveness of this technique for the characterization of urban environments.
{"title":"Three-Dimensional Urban Characterization Using Polarimetric SAR Correlation Tomographic Techniques and TSX/TDX Images","authors":"Xing Peng, Yue Huang, L. Ferro-Famil, Jianjun Zhu, Yanan Du, Haiqiang Fu","doi":"10.1109/IGARSS.2019.8900289","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8900289","url":null,"abstract":"Polarimetric synthetic aperture radar tomography (Pol-TomoSAR) allows to achieve a 3-D characterization over urban areas using multiple polarimetric acquisitions. However, using spaceborne datasets, such as TerraSAR-X, it is difficult to localize the distributed or uncorrelated scattering patterns along elevation due to the temporal decorrelation. In order to overcome this limitation, this paper proposes polarimetric correlation tomographic techniques based on Tandem-mode images. The key of this technique is to build a covariance matrix from the observed Tandem coherence pairs, and then apply conventional covariance-based tomographic techniques. This processing allows to extract both coherent and distributed scatterers. The resulting 3-D reconstruction is more refined and detailed, compared to the one derived from TerraSAR-X data. Seven TSX/TDX pairs in fully polarimetric mode over a small county in Yunnan province, China, are used to demonstrate the effectiveness of this technique for the characterization of urban environments.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"26 1","pages":"4924-4926"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73081810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8900388
Zheng Zhang, Changmiao Hu, Ping Tang, T. Corpetti
Cloud-contaminated pixels exist ubiquitously in satellite images, which limit the usability of satellite images and increase the difficulty of image analysis. To reconstruct these pixels, a basic idea is to transfer cloud-free pixels from corresponding multi-temporal images to the target image, and the performance of this category of methods depends on the quality of information transfer between images. We propose in this work a novel pixel reconstruction method based on optimal transport. Our method first conducts an adaptive col-or transfer between multi-temporal images and then replaces cloud-contaminated pixels by transferred cloud-free pixels. The proposed method fully explores the potential of optimal transport to generate a more adaptive color transfer plan and thus ensure a high quality information transfer between images. Compared with other widely used methods, visual and statistical results on Landsat and MODIS images demonstrate the capacity of our method.
{"title":"Color Adaptation and Cloud Removal between Satellite Images via Optimal Transport","authors":"Zheng Zhang, Changmiao Hu, Ping Tang, T. Corpetti","doi":"10.1109/IGARSS.2019.8900388","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8900388","url":null,"abstract":"Cloud-contaminated pixels exist ubiquitously in satellite images, which limit the usability of satellite images and increase the difficulty of image analysis. To reconstruct these pixels, a basic idea is to transfer cloud-free pixels from corresponding multi-temporal images to the target image, and the performance of this category of methods depends on the quality of information transfer between images. We propose in this work a novel pixel reconstruction method based on optimal transport. Our method first conducts an adaptive col-or transfer between multi-temporal images and then replaces cloud-contaminated pixels by transferred cloud-free pixels. The proposed method fully explores the potential of optimal transport to generate a more adaptive color transfer plan and thus ensure a high quality information transfer between images. Compared with other widely used methods, visual and statistical results on Landsat and MODIS images demonstrate the capacity of our method.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"36 1","pages":"787-790"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78301615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-28DOI: 10.1109/IGARSS.2019.8898563
R. C. Daudt, Adrien Chan-Hon-Tong, B. L. Saux, Alexandre Boulch
In this paper we discuss the issues of using inexact and inaccurate ground truth in the context of supervised learning. To leverage large amounts of Earth observation data for training algorithms, one often has to use ground truth which was not been carefully assessed. We address both the problems of training and evaluation. We first propose a weakly supervised approach for training change classifiers which is able to detect pixel-level changes in aerial images. We then propose a data poisoning approach to get a reliable estimate of the accuracy that can be expected from a classifier, even when the only ground-truth available does not match the reality. Both are assessed on practical land use and land cover applications.
{"title":"Learning to Understand Earth Observation Images with Weak and Unreliable Ground Truth","authors":"R. C. Daudt, Adrien Chan-Hon-Tong, B. L. Saux, Alexandre Boulch","doi":"10.1109/IGARSS.2019.8898563","DOIUrl":"https://doi.org/10.1109/IGARSS.2019.8898563","url":null,"abstract":"In this paper we discuss the issues of using inexact and inaccurate ground truth in the context of supervised learning. To leverage large amounts of Earth observation data for training algorithms, one often has to use ground truth which was not been carefully assessed. We address both the problems of training and evaluation. We first propose a weakly supervised approach for training change classifiers which is able to detect pixel-level changes in aerial images. We then propose a data poisoning approach to get a reliable estimate of the accuracy that can be expected from a classifier, even when the only ground-truth available does not match the reality. Both are assessed on practical land use and land cover applications.","PeriodicalId":13262,"journal":{"name":"IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium","volume":"13 1","pages":"5602-5605"},"PeriodicalIF":0.0,"publicationDate":"2019-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85247643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：