Using deep learning to classify grassland management intensity in ground-level photographs for more automated production of satellite land use maps

Saadeldin, Mohamed; O’Hara, Robert B.; Zimmermann, Jesko; Namee, Brian Mac; Green, Stuart

doi:10.1016/j.rsase.2022.100741

Cited by 9 publications

(5 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A comparison of Figures 4 and 5 also demonstrates that there is little practical difference between the scatter plots of STR-EVI and STR-MSAVI at low values of vegetation cover, probably because grass is present at all stages of the agricultural/cattle grazing cycle for the study areas. The study also corroborates results from Sentinel-1 and S-2 imagery which indicate that EVI is optimum for the classification of Irish grassland [65].…”

Section: Edge Curves and Vegetation Indexsupporting

confidence: 84%

Temporal Stability of Grassland Soil Moisture Utilising Sentinel-2 Satellites and Sparse Ground-Based Sensor Networks

Basu,

Daly,

Brown

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Soil moisture is important for understanding climate, water resources, water storage, and land use management. This study used Sentinel-2 (S-2) satellite optical data to retrieve surface soil moisture at a 10 m scale on grassland sites with low hydraulic conductivity soil in a climate dominated by heavy rainfall. Soil moisture was estimated after modifying the Optical Trapezoidal Model to account for mixed land cover in such conditions. The method uses data from a short-wave infra-red band, which is sensitive to soil moisture, and four vegetation indices from optical bands, which are sensitive to overlying vegetation. Scatter plots of these data from multiple, infrequent satellite passes are used to define the range of surface moisture conditions. The saturated and dry edges are clearly non-linear, regardless of the choice of vegetation index. Land cover masks are used to generate scatter plots from data only over grassland sites. The Enhanced Vegetation Index demonstrated advantages over other vegetation indices for surface moisture estimation over the entire range of grassland conditions. In poorly drained soils, the time lag between satellite surface moisture retrievals and in situ sensor soil moisture at depth must be part of the validation process. This was achieved by combining an approximate solution to the Richards’ Equation, along with measurements of saturated and residual moisture from soil samples, to optimise the correlations between measurements from satellites and sensors at a 15 cm depth. Time lags of 2–4 days resulted in a reduction of the root mean square errors between volumetric soil moisture predicted from S-2 data and that measured by in situ sensors, from ~0.1 m3/m3 to <0.06 m3/m3. The surface moisture results for two grassland sites were analysed using statistical concepts based upon the temporal stability of soil water content, an ideal framework for the intermittent Sentinel-2 data in conditions of persistent cloud cover. The analysis could discriminate between different natural drainages and surface soil textures in grassland areas and could identify sub-surface artificial drainage channels. The techniques are transferable for land-use and agricultural management in diverse environmental conditions without the need for extensive and expensive in situ sensor networks.

show abstract

Section: Edge Curves and Vegetation Indexsupporting

confidence: 84%

Temporal Stability of Grassland Soil Moisture Utilising Sentinel-2 Satellites and Sparse Ground-Based Sensor Networks

Basu,

Daly,

Brown

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The present study has some limitations regarding species classification, which may be improved by combining deep learning and spectral features to increase accuracy and efficiency (Saadeldin et al, 2022) such as the integration of vegetation height and canopy extraction and species classification. Further in‐depth research could also explore the effects of lower flight altitudes (e.g., 10–50 m) and adjusted flight attitudes (e.g., oblique photogrammetry) for more accurate monitoring of low‐growing herbaceous or recently emerged seedlings.…”

Section: Discussionmentioning

confidence: 99%

“…The present study has some limitations regarding species classification, which may be improved by combining deep learning and spectral features to increase accuracy and efficiency (Saadeldin et al, 2022) Remote sensing provides basic variables for assessing and monitoring diversity, and establishing relationships between plant diversity and spectral data has been proposed as a potential solution (Chapungu et al, 2020;Gholizadeh et al, 2020). The NDVI is frequently used to assess vegetation health, greenness, and estimate vegetation species diversity.…”

Section: Feasibility Of Vegetation Species Identificationmentioning

confidence: 99%

Using unmanned aerial vehicle multispectral data for monitoring the outcomes of ecological restoration in mining areas

Chen,

Hou,

Zhang

et al. 2023

Land Degrad Dev

View full text Add to dashboard Cite

The effective and efficient monitoring of revegetation outcomes is a key component of ecosystem restoration. Monitoring often involves labor‐intensive manual methods, which are difficult to deploy when sites are inaccessible or involve large areas of revegetation. This study aimed to identify plant species and quantify α‐diversity index on a sub‐meter scale at Manlailiang Mine Site in Northwestern China using unmanned aerial vehicles (UAVs) as a means to semiautomate large‐scale vegetation monitoring. UAVs equipped with multispectral sensors were combined with three industry‐standard supervised classification algorithms (support vector machine [SVM], maximum likelihood, and artificial neural network) to classify plant species. Spectral vegetation indices (normalized difference vegetation index [NDVI], difference vegetation index [DVI], visible‐band difference vegetation index, soil‐adjusted vegetation index, modified soil‐adjusted vegetation index, and excess green–excess red) were used to assess vegetation diversity obtained from on‐ground survey plot data (the Margalef, Pielou, Simpson, and Shannon–Wiener indices). Our results showed that SVM outperformed other algorithms in species identification accuracy (overall accuracy of 84%). Significant relationships were observed between vegetation indices and diversity indices, with the DVI performing significantly better than many more commonly used indices such as the NDVI. The findings highlight the potential of combining UAV multispectral data, spectral vegetation indices and ground surveys for effective and efficient fine‐scale monitoring of vegetation diversity in the ecological restoration of mining areas. This has significant practical benefits for improving adaptive management of restoration through improved monitoring tools.

show abstract

“…Additionally, another modification made to OPTRAM in this study is the use of the Enhanced Vegetation index (EVI) instead of the Normalised Difference Vegetation Index (NDVI) as used originally in OPTRAM. EVI has several advantages over NDVI especially with respect to saturation issues over vegetated lands, where its sensitivity decreases with increasing biomass (Rocha and Shaver, 2009;Antunes Daldegan et al, 2020;Ojha et al, 2021) and based on class separability, it has also been shown to be an optimum VI for studying Irish grassland using S-2 data (Saadeldin et al, 2022). nSSM maps for the farm were obtained at a spatial resolution of 10 m (Figure 2).…”

Section: Optrammentioning

confidence: 99%

Identifying favourable conditions for farm scale trafficability and grass growth using a combined Sentinel-2 and soil moisture deficit approach

Basu,

Fenton,

Daly

et al. 2024

Front. Environ. Sci.

View full text Add to dashboard Cite

In Atlantic Europe, on poorly drained grasslands soils, compaction negatively affects soil health when trafficked in wet conditions, while optimum grass growth cannot be achieved in excessively dry conditions. In Ireland, daily soil moisture deficit (SMD) information is forecasted at regional scale for all soil drainage classes. Optimal paddock conditions can occur between trafficking (10 mm) and optimum grass growth (50 mm) SMD thresholds for an identified drainage class. The objective of this farm scale study is to improve the identification of optimum conditions in time and space by combining high resolution spatial soil moisture estimates with soil drainage class specific SMD data. For that purpose, Sentinel- 2 (S-2) data was used in a modified Optical Trapezoid Model (OPTRAM) to derive normalised surface soil moisture (nSSM) estimates at farm level. In-situ soil moisture sensors providing daily estimates of volumetric soil moisture were used for validation of OPTRAM with an RMSE of 0.05. Cumulative 7-day SMD prior to the date of each S-2 image was analysed for each year from 2017-2021 to select nSSM maps corresponding to negative, 0 or −0 and positive SMD. Results established a relationship between nSSM and SMD indicating optimal conditions changed spatially and temporally. The months of April, May, August and September always presented at least 35% of the farm area available for optimum management operations. Future refinement of this methodology utilising daily high resolution remote sensing data could provide near real-time information for farmers.

show abstract

Using deep learning to classify grassland management intensity in ground-level photographs for more automated production of satellite land use maps

Cited by 9 publications

References 44 publications

Temporal Stability of Grassland Soil Moisture Utilising Sentinel-2 Satellites and Sparse Ground-Based Sensor Networks

Temporal Stability of Grassland Soil Moisture Utilising Sentinel-2 Satellites and Sparse Ground-Based Sensor Networks

Using unmanned aerial vehicle multispectral data for monitoring the outcomes of ecological restoration in mining areas

Identifying favourable conditions for farm scale trafficability and grass growth using a combined Sentinel-2 and soil moisture deficit approach

Contact Info

Product

Resources

About