The form of a remote sensing index is generally empirically defined, whether by choosing specific reflectance bands, equation forms or its coefficients. These spectral indices are used as preprocessing stage before object detection/classification. But no study seems to search for the best form through function approximation in order to optimize the classification and/or segmentation. The objective of this study is to develop a method to find the optimal index, using a statistical approach by gradient descent on different forms of generic equations. From six wavebands images, five equations have been tested, namely: linear, linear ratio, polynomial, universal function approximator and dense morphological. Few techniques in signal processing and image analysis are also deployed within a deep-learning framework. Performances of standard indices and DeepIndices were evaluated using two metrics, the dice (similar to f1-score) and the mean intersection over union (mIoU) scores. The study focuses on a specific multispectral camera used in near-field acquisition of soil and vegetation surfaces. These DeepIndices are built and compared to 89 common vegetation indices using the same vegetation dataset and metrics. As an illustration the most used index for vegetation, NDVI (Normalized Difference Vegetation Indices) offers a mIoU score of 63.98% whereas our best models gives an analytic solution to reconstruct an index with a mIoU of 82.19%. This difference is significant enough to improve the segmentation and robustness of the index from various external factors, as well as the shape of detected elements.
", "keywords": ["multi-spectral", "[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing", "multispectral", "Science", "0211 other engineering and technologies", "[SDV.SA.STA] Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "02 engineering and technology", "Spectral indice", "Deep-learning", "image; precision agriculture; spectral indices; multi-spectral; deep-learning; vegetation segmentation", "deep-learning", "[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing", "[SDV.SA.STA]Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "[SDV.BV]Life Sciences [q-bio]/Vegetal Biology", "[SDV.BV] Life Sciences [q-bio]/Vegetal Biology", "image", "precision agriculture", "Precision agriculture", "Vegetation segmentation", "Multi-spectral", "Q", "04 agricultural and veterinary sciences", "15. Life on land", "004", "Image", "vegetation segmentation", "spectral indices", "0401 agriculture", " forestry", " and fisheries"]}, "links": [{"href": "http://www.mdpi.com/2072-4292/13/12/2261/pdf"}, {"href": "https://www.mdpi.com/2072-4292/13/12/2261/pdf"}, {"href": "https://doi.org/10.3390/rs13122261"}, {"rel": "related", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/Remote%20Sensing", "name": "related record", "description": "related record", "type": "application/json"}, {"rel": "self", "type": "application/geo+json", "title": "10.3390/rs13122261", "name": "item", "description": "10.3390/rs13122261", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/10.3390/rs13122261"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2021-06-09T00:00:00Z"}}, {"id": "2809041101", "type": "Feature", "geometry": null, "properties": {"updated": "2026-07-26T16:25:49Z", "type": "Journal Article", "created": "2018-06-18", "title": "Estimating the water budget components of irrigated crops: Combining the FAO-56 dual crop coefficient with surface temperature and vegetation index data", "description": "Abstract The FAO-56 dual crop coefficient (FAO-2Kc) model has been extensively used at the field scale to estimate the crop water requirements by means of the simulated evapotranspiration (ET) and its two components evaporation (E) and transpiration (T). Given that the main limitation of FAO-2Kc for operational irrigation management over large areas is the unavailability (over most irrigated areas) of irrigation data, this study investigates the feasibility 1) to constrain the FAO-2Kc ET from LST and VI data, 2) to retrieve irrigation amounts and dates from LST and VI data and 3) to estimate the root-zone soil moisture (RZSM) at the daily scale. In practice, the vegetation and soil temperatures retrieved from LST/VI data are used to estimate the FAO-2Kc vegetation stress coefficient (Ks) and soil evaporation reduction coefficient (Kr), respectively. The modeling and remote sensing combined approach is tested over a wheat crop field in central Morocco, and results are evaluated in terms of ET, irrigation and RZSM estimates. ET is estimated with a RMSE of 0.68\u202fmm day-1 compared to 0.84\u202fmm day-1 for the standard (without using LST data) FAO-2Kc based on tabulated values for the parameters. The total irrigation depth (67\u202fmm) is correctly estimated and is very close to the actual effective irrigation (69.8\u202fmm) applied by the farmer. Daily RZSM is estimated with an R2 value of 0.68 (0.42) and a RMSE value of 0.034 (0.061) m3 m-3 by forcing FAO-2Kc using the retrieved irrigation (from LST-derived estimates and precipitation only). Since spaceborne LST data are currently not available at both high-spatial and high-temporal resolution, a sensitivity analysis is finally undertaken to assess the potential and applicability of the proposed methodology to temporally-sparse thermal data.", "keywords": ["FAO-56", "0106 biological sciences", "2. Zero hunger", "550", "Evapotranspiration", "[SDE.IE]Environmental Sciences/Environmental Engineering", "Root-zone soil moisture", "[SDV.SA.STA] Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "Root-Zone Soil Moisture", "Surface Temperature", "[INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation", "01 natural sciences", "6. Clean water", "Surface temperature", "[SDV.SA.STA]Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "[INFO.INFO-MO] Computer Science [cs]/Modeling and Simulation", "[SDE.IE] Environmental Sciences/Environmental Engineering", "Irrigation", "0105 earth and related environmental sciences"]}, "links": [{"href": "https://doi.org/2809041101"}, {"rel": "related", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/Agricultural%20Water%20Management", "name": "related record", "description": "related record", "type": "application/json"}, {"rel": "self", "type": "application/geo+json", "title": "2809041101", "name": "item", "description": "2809041101", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/2809041101"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2018-09-01T00:00:00Z"}}, {"id": "10.5220/0009169301030110", "type": "Feature", "geometry": null, "properties": {"updated": "2026-07-26T16:21:53Z", "type": "Journal Article", "created": "2020-03-19", "title": "Two-step Multi-spectral Registration Via Key-point Detector and Gradient Similarity: Application to Agronomic Scenes for Proxy-sensing", "description": "The potential of multi-spectral images is growing rapidly in precision agriculture, and is currently based on the use of multi-sensor cameras. However, their development usually concerns aerial applications and their parameters are optimized for high altitudes acquisition by drone (UAV \u2248 50 meters) to ensure surface coverage and reduce technical problems. With the recent emergence of terrestrial robots (UGV), their use is diverted for nearby agronomic applications. Making it possible to explore new agronomic applications, maximizing specific traits extraction (spectral index, shape, texture \u2026) which requires high spatial resolution. The problem with these cameras is that all sensors are not aligned and the manufacturers\u2019 methods are not suitable for close-field acquisition, resulting in offsets between spectral images and degrading the quality of extractable informations. We therefore need a solution to accurately align images in such condition. In this study we propose a two-steps method applied to the six-bands Airphen multi-sensor camera with (i) affine correction using pre-calibrated matrix at different heights, the closest transformation can be selected via internal GPS and (ii) perspective correction to refine the previous one, using key-points matching between enhanced gradients of each spectral bands. Nine types of key-point detection algorithms (ORB, GFTT, AGAST, FAST, AKAZE, KAZE, BRISK, SURF, MSER) with three different modalities of parameters were evaluated on their speed and performances, we also defined the best reference spectra on each of them. The results show that GFTT is the most suitable methods for key-point extraction using our enhanced gradients, and the best spectral reference was identified to be the band centered on 570 nm for this one. Without any treatment the initial error is about 62 px, with our method, the remaining residual error is less than 1 px, where the manufacturer\u2019s involves distortions and loss of information with an estimated residual error of approximately 12 px", "keywords": ["03 medical and health sciences", "0302 clinical medicine", "Registration", "Registration", " Multi-spectral imagery", " Precision farming", " Feature descriptor", "0202 electrical engineering", " electronic engineering", " information engineering", "[SDV.SA.STA] Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "Precision farming", "Feature descriptor", "Multi-spectral imagery", "02 engineering and technology", "15. Life on land", "[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing"]}, "links": [{"href": "https://doi.org/10.5220/0009169301030110"}, {"rel": "related", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/Proceedings%20of%20the%2015th%20International%20Joint%20Conference%20on%20Computer%20Vision%2C%20Imaging%20and%20Computer%20Graphics%20Theory%20and%20Applications", "name": "related record", "description": "related record", "type": "application/json"}, {"rel": "self", "type": "application/geo+json", "title": "10.5220/0009169301030110", "name": "item", "description": "10.5220/0009169301030110", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/10.5220/0009169301030110"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2020-01-01T00:00:00Z"}}, {"id": "10468/11557", "type": "Feature", "geometry": null, "properties": {"updated": "2026-07-26T16:24:46Z", "type": "Report", "title": "SARMENTI: Smart multisensor embedded and secure system for soil nutrient and gaseous emission monitoring", "description": "Demand for sustainably produced food is driving current strategies for intensification of the agricultural sector worldwide. To meet these challenges farmers will need to adopt a whole-farm approach to resource efficiency. They will increase their productivity with a better application of knowledge per hectare. Optimising soil fertility will enable farmers to maximise their productivity and profitability with higher grass and crop yield and quality.", "keywords": ["[INFO.INFO-NI] Computer Science [cs]/Networking and Internet Architecture [cs.NI]", "Food", "[SDV.SA.STA] Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "Agriculture", "Soil fertility", "[SPI.TRON] Engineering Sciences [physics]/Electronics", "Resource efficiency"], "contacts": [{"organization": "Lesecq, Suzanne, Molnos, A., Gougis, M., Gouze, E., Correvon, M., Dudnik, G., Di Matteo, A., Di Palma, V., Alessi, E., O'Riordan, A., \u00d3'Murch\u00fa, C., Lovera, P., Ponsardin, G., Lapierre, E., Daly, K., Balan, Titus, Dimitru, C., Mailat, G.,", "roles": ["creator"]}]}, "links": [{"href": "https://doi.org/10468/11557"}, {"rel": "self", "type": "application/geo+json", "title": "10468/11557", "name": "item", "description": "10468/11557", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/10468/11557"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2019-01-01T00:00:00Z"}}, {"id": "2078.1/260550", "type": "Feature", "geometry": null, "properties": {"updated": "2026-07-26T16:25:30Z", "type": "Journal Article", "created": "2022-05-02", "title": "Mineral element recycling in topsoil following permafrost degradation and a vegetation shift in sub-Arctic tundra", "description": "Climate change affects the Arctic and sub-Arctic regions by exposing previously frozen permafrost to thaw, unlocking soil nutrients, changing hydrological processes, and boosting plant growth. As a result, sub-Arctic tundra is subject to a shrub expansion, called \u201cshrubification\u201d, at the expense of sedge species. Depending on the intrinsic foliar properties of these plant species, changes in foliar mineral element fluxes with shrubification in the context of permafrost degradation may influence topsoil mineral element composition. Despite the potential implications of changes in topsoil mineral element concentrations for the fate of organic carbon, this remains poorly quantified. Here, we investigate vegetation foliar and topsoil mineral element composition (Si, K, Ca, P, Mn, Zn, Cu, Mo, V) across a natural gradient of permafrost degradation at a typical sub-Arctic tundra at Eight Mile Lake (Alaska, USA). Results show that foliar mineral element concentrations are higher (up to 9 times; Si, K, Mo for all species, and for some species Zn) or lower (up to 2 times; Ca, P, Mn, Cu, V for all species, and for some species Zn) in sedge than in shrub species. As a result, a vegetation shift over ~40 years has resulted in lower topsoil concentrations in Si, K, Zn, and Mo (respectively of 52, 24, 20, and 51%) in highly degraded permafrost sites compared to poorly degraded permafrost sites due to lower foliar fluxes of these elements. For other elements (Ca, P, Mn, Cu, and V), the vegetation shift has not induced a marked change in topsoil concentrations at this current stage of permafrost degradation. A modeled amplified shrubification associated with a further permafrost degradation is expected to increase foliar Ca, P, Mn, Cu, and V fluxes, which will likely change these element concentrations in topsoil. These data can serve as a first estimate to assess the influence of other shifts in vegetation in Arctic and sub-Arctic tundra such as sedge expansion under wetter soil conditions.", "keywords": ["topsoil", "[SDV.SA.STA] Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "mineral elements", "04 agricultural and veterinary sciences", "sub-Arctic tundra", "15. Life on land", "01 natural sciences", "vegetation change", "13. Climate action", "[SDV.SA.STA]Life Sciences [q-bio]/Agricultural sciences/Sciences and technics of agriculture", "0401 agriculture", " forestry", " and fisheries", "shrubification", "permafrost degradation", "0105 earth and related environmental sciences"]}, "links": [{"href": "https://doi.org/2078.1/260550"}, {"rel": "related", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/Geoderma", "name": "related record", "description": "related record", "type": "application/json"}, {"rel": "self", "type": "application/geo+json", "title": "2078.1/260550", "name": "item", "description": "2078.1/260550", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/2078.1/260550"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2022-09-01T00:00:00Z"}}, {"id": "2164/19434", "type": "Feature", "geometry": null, "properties": {"updated": "2026-07-26T16:25:36Z", "type": "Journal Article", "created": "2022-08-22", "title": "Global systematic review with meta-analysis reveals yield advantage of legume-based rotations and its drivers", "description": "AbstractDiversified cropping systems, especially those including legumes, have been proposed to enhance food production with reduced inputs and environmental impacts. However, the impact of legume pre-crops on main crop yield and its drivers has never been systematically investigated in a global context. Here, we synthesize 11,768 yield observations from 462 field experiments comparing legume-based and non-legume cropping systems and show that legumes enhanced main crop yield by 20%. These yield advantages decline with increasing N fertilizer rates and crop diversity of the main cropping system. The yield benefits are consistent among main crops (e.g., rice, wheat, maize) and evident across pedo-climatic regions. Moreover, greater yield advantages (32% vs. 7%) are observed in low- vs. high-yielding environments, suggesting legumes increase crop production with low inputs (e.g., in Africa or organic agriculture). In conclusion, our study suggests that legume-based rotations offer a critical pathway for enhancing global crop production, especially when integrated into low-input and low-diversity agricultural systems.