PhotonLabeler: An Inter-Disciplinary Platform for Visual Interpretation and Labeling of ICESat-2 Geolocated Photon Data

Malambo, Lonesome; Popescu, Sorin C.

doi:10.3390/rs12193168

Cited by 12 publications

(16 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ATL08 95th and 98th percentile canopy heights were more reliable than intermediate (25th-50th) height percentiles, which makes ATL08 data suited for routine canopy height estimation as applied in this study. Our other study also assessed the quality of ATL08 photon classifications over this study and a site in Zambia showing high agreement between ATL08 photon classification and manually labeled data [24]. These observations provide a reasonable basis for applying ATL08 data, but selection of data granules to avoid data corrupted by instrument errors or sub-optimal photon classification is still vital for reliable canopy height modeling [24].…”

Section: Icesat-2 Datamentioning

confidence: 60%

Landsat-Scale Regional Forest Canopy Height Mapping Using ICESat-2 Along-Track Heights: Case Study of Eastern Texas

Malambo

Popescu

2022

Remote Sensing

View full text Add to dashboard Cite

Spaceborne profiling lidar missions such as the Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) are collecting direct elevation measurements, supporting the retrieval of vegetation attributes such as canopy height that are crucial in forest carbon and ecological studies. However, such profiling lidar systems collect observations along predefined ground tracks which limit the spatially complete mapping of forest canopy height. We demonstrate that the fusion of ICESat-2 along-track canopy height estimates and ancillary Landsat and LANDFIRE (Landscape Fire and Resource Management Planning Tools Project) data can enable the generation of spatially complete canopy height data at a regional level in the United States. We developed gradient-boosted regression models relating canopy heights with ancillary data values and used them to predict canopy height in unobserved locations at a 30 m spatial resolution. Model performance varied (R2 = 0.44 − 0.50, MAE = 2.61–2.80 m) when individual (per month) Landsat data and LANDFIRE data were used. Improved performance was observed when combined Landsat and LANDFIRE data were used (R2 = 0.69, MAE = 2.09 m). We produced a gridded canopy height product over our study area in eastern Texas, which agreed moderately (R2 = 0.46, MAE = 4.38 m) with independent airborne lidar-derived canopy heights. Further, we conducted a comparative assessment with the Global Forest Canopy Height product, an existing 30 m spatial resolution canopy height product generated using GEDI (Global Ecosystem Dynamics Investigation) canopy height and multitemporal Landsat data. In general, our product showed better agreement with airborne lidar heights than the global dataset (R2 = 0.19 MAE = 5.83 m). Major differences in canopy height values between the two products are attributed to land cover changes, height metrics used (98th in this study vs 95th percentile), and the inherent differences in lidar sampling and their geolocation uncertainties between ICESat-2 and GEDI. On the whole, our integration of ICESat-2 data with ancillary datasets was effective for spatially complete canopy height mapping. For better modeling performance, we recommend the careful selection of ICESat-2 datasets to remove erroneous data and applying a series of Landsat data to account for phenological changes. The canopy height product provides a valuable spatially detailed and synoptic view of canopy heights over the study area, which would support various forestry and ecological assessments at an enhanced 30 Landsat spatial resolution.

show abstract

Section: Icesat-2 Datamentioning

confidence: 60%

Landsat-Scale Regional Forest Canopy Height Mapping Using ICESat-2 Along-Track Heights: Case Study of Eastern Texas

Malambo

Popescu

2022

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The proposed terrain profile extraction method was tested using ICESat-2 s data measured by the weak beam in four mountainous areas, where the classical DBSCAN algorithm cannot successfully detect the signal photons from weak beam data. Both the classical DBSCAN algorithm and the modified DBSCAN algorithm in this study were separately applied to extract terrain profile, and the results were compared with the truth-values, which were obtained by labeling the signal photons manually [30,35,38]. As above-mentioned, in the classification result provided by the ATL03 dataset, geolocated photons were marked with different confidence levels (i.e., 0 = noise; 1 = added to allow for buffer but algorithm classifies as background; 2 = low; 3 = medium; 4 = high) [42], and normally, photons with a confidence level larger than 1 can be considered as signal photons.…”

Section: Resultsmentioning

confidence: 99%

“…However, the method of this study focused on remote mountainous areas with large slopes, where few in situ measurements are available. As a result, similar to many previous studies [30,35,38], the visual inspection is used as an evaluation standard for photon-counting cloud filtering when the in situ ground-truth is unavailable.…”

Section: (5) Running a 3σ Confidence Filter To Remove Outliersmentioning

confidence: 96%

“…As the measured photons from profiling photon-counting LiDARs are quite different, NASA's research team has proposed many specific surface-finding algorithms, which performed well on, planned initially to some specific surface types (e.g., sea ice, ice sheets, land/vegetation, ocean, and inland water) [17][18][19][20][21][22][23][24][25][26][27][28][29][30]. The terrain characteristics significantly vary in different land-cover types (e.g., with different reflectivity, roughness, and slopes) and make the distribution of signal photons quite different.…”

Section: Introductionmentioning

confidence: 99%

“…DRAGANN algorithms have been used to process ATL03 data and generate the mission's land-vegetation along-track products (ATL08) [27]. Malambo et al [30] developed an inter-disciplinary platform, named PhotonLabeler, for visual interpretation and labeling for the ICESat-2 data product. This platform can be used to provide training and validation data in the development of new algorithms for the ICESat-2 mission.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Signal Photon Extraction Method for Weak Beam Data of ICESat-2 Using Information Provided by Strong Beam Data in Mountainous Areas

Zhang

Liu

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

The Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) can measure the elevations of the Earth’s surface using a sampling strategy with unprecedented spatial detail. In the daytime of mountainous areas where the signal–noise ratio (SNR) of weak beam data is very low, current algorithms do not always perform well on extracting signal photons from weak beam data (i.e., many signal photons were missed). This paper proposes an effective algorithm to extract signal photons from the weak beam data of ICESat-2 in mountainous areas. First, a theoretical equation of SNR for ICESat-2 measured photons in mountainous areas was derived to prove that the available information provided by strong beam data can be used to assist the signal extraction of weak beam data (that may have very low SNR in mountainous areas). Then, the relationship between the along-track slope and the noise level was used as the bridge to connect the strong and weak beam data. To be specific, the along-track slope of the weak beam was inversed by the slope–noise relationship obtained from strong beam data, and then was used to rotate the direction of the searching neighborhood in the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. With the help of this process, the number of signal photons included in the searching neighborhood will significantly increase in mountainous areas and will be easily detected from the measured noisy photons. The proposed algorithm was tested in the Tibetan Plateau, the Altun Mountains, and the Tianshan Mountains in different seasons, and the extraction results were compared with the results from the ATL03 datasets, the ATL08 datasets, and the classical DBSCAN algorithm. Based on the ground-truth signal photons obtained by visual inspection, the parameters of the classification precision, recall, and F-score of our algorithm and three other algorithms were calculated. The modified DBSCAN could achieve a good balance between the classification precision (93.49% averaged) and recall (89.34% averaged), and its F-score (more than 0.91) was higher than that of the other three methods, which successfully obtained a continuous surface profile from weak beam data with very low SNRs. In the future, the detected signal photons from weak beam data are promising to assess the elevation accuracy achieved by ICESat-2, estimate the along-track and cross-track slope, and further obtain the ground control points (GCPs) for stereo-mapping satellites in mountainous areas.

show abstract

Land-sea classification based on the fast feature detection model for ICESat-2 ATL03 datasets

Li,

Chu,

et al. 2024

International Journal of Applied Earth Observation and Geoinfor

View full text Add to dashboard Cite

PhotonLabeler: An Inter-Disciplinary Platform for Visual Interpretation and Labeling of ICESat-2 Geolocated Photon Data

Cited by 12 publications

References 20 publications

Landsat-Scale Regional Forest Canopy Height Mapping Using ICESat-2 Along-Track Heights: Case Study of Eastern Texas

Landsat-Scale Regional Forest Canopy Height Mapping Using ICESat-2 Along-Track Heights: Case Study of Eastern Texas

Signal Photon Extraction Method for Weak Beam Data of ICESat-2 Using Information Provided by Strong Beam Data in Mountainous Areas

Land-sea classification based on the fast feature detection model for ICESat-2 ATL03 datasets

Contact Info

Product

Resources

About