Records 1 - 20 / 361
Comprehensive mass spectrometry-guided phenotyping of plant specialized metabolites reveals metabolic diversity in the cosmopolitan plant family Rhamnaceae
Kang, Kyo Bin ; Ernst, Madeleine ; Hooft, Justin J.J. van der; Silva, Ricardo R. da; Park, Junha ; Medema, Marnix H. ; Sung, Sang Hyun ; Dorrestein, Pieter C. - \ 2019
The Plant Journal (2019). - ISSN 0960-7412
annotation - classification - mass spectrometry - Rhamnaceae - specialized metabolites - technical advance
Plants produce a myriad of specialized metabolites to overcome their sessile habit and combat biotic as well as abiotic stresses. Evolution has shaped the diversity of specialized metabolites, which then drives many other aspects of plant biodiversity. However, until recently, large-scale studies investigating the diversity of specialized metabolites in an evolutionary context have been limited by the impossibility of identifying chemical structures of hundreds to thousands of compounds in a time-feasible manner. Here we introduce a workflow for large-scale, semi-automated annotation of specialized metabolites and apply it to over 1000 metabolites of the cosmopolitan plant family Rhamnaceae. We enhance the putative annotation coverage dramatically, from 2.5% based on spectral library matches alone to 42.6% of total MS/MS molecular features, extending annotations from well-known plant compound classes into dark plant metabolomics. To gain insights into substructural diversity within this plant family, we also extract patterns of co-occurring fragments and neutral losses, so-called Mass2Motifs, from the dataset; for example, only the Ziziphoid clade developed the triterpenoid biosynthetic pathway, whereas the Rhamnoid clade predominantly developed diversity in flavonoid glycosides, including 7-O-methyltransferase activity. Our workflow provides the foundations for the automated, high-throughput chemical identification of massive metabolite spaces, and we expect it to revolutionize our understanding of plant chemoevolutionary mechanisms.
Potato Virus Y Detection in Seed Potatoes Using Deep Learning on Hyperspectral Images
Polder, G. ; Blok, P.M. ; Villiers, H.A.C. de; Wolf, J.M. van der; Kamp, J.A.L.M. - \ 2019
Frontiers in Plant Science 10 (2019). - ISSN 1664-462X
crop resistance - Phenotyping - hyperspectral imaging - classification - Convolutional neural network - Solanum tuberosum
Virus diseases are of high concern in the cultivation of seed potatoes. Once found inthe field, virus diseased plants lead to declassification or even rejection of the seed lotsresulting in a financial loss. Farmers put in a lot of effort to detect diseased plants andremove virus-diseased plants from the field. Nevertheless, dependent on the cultivar,virus diseased plants can be missed during visual observations in particular in an earlystage of cultivation. Therefore, there is a need for fast and objective disease detection.Early detection of diseased plants with modern vision techniques can significantlyreduce costs. Laboratory experiments in previous years showed that hyperspectral imaging clearly could distinguish healthy from virus infected potato plants. This paper reports on our first real field experiment. A new imaging setup was designed, consisting of a hyperspectral line-scan camera. Hyperspectral images were taken in the field with a line interval of 5 mm. A fully convolutional neural network was adapted for hyperspectral images and trained on two experimental rows in the field. The trained network was validated on two other rows, with different potato cultivars. For three of the four row/date combinations the precision and recall compared to conventional disease assessment exceeded 0.78 and 0.88, respectively. This proves the suitability of this method for real world disease detection.
Dense semantic labeling of subdecimeter resolution images with convolutional neural networks
Volpi, Michele ; Tuia, Devis - \ 2017
IEEE Transactions on Geoscience and Remote Sensing 55 (2017)2. - ISSN 0196-2892 - p. 881 - 893.
Aerial images - classification - convolutional neural networks (CNNs) - deconvolution networks - deep learning - semantic labeling - subdecimeter resolution
Semantic labeling (or pixel-level land-cover classification) in ultrahigh-resolution imagery (<10 cm) requires statistical models able to learn high-level concepts from spatial data, with large appearance variations. Convolutional neural networks (CNNs) achieve this goal by learning discriminatively a hierarchy of representations of increasing abstraction. In this paper, we present a CNN-based system relying on a downsample-then-upsample architecture. Specifically, it first learns a rough spatial map of high-level representations by means of convolutions and then learns to upsample them back to the original resolution by deconvolutions. By doing so, the CNN learns to densely label every pixel at the original resolution of the image. This results in many advantages, including: 1) the state-of-the-art numerical accuracy; 2) the improved geometric accuracy of predictions; and 3) high efficiency at inference time. We test the proposed system on the Vaihingen and Potsdam subdecimeter resolution data sets, involving the semantic labeling of aerial images of 9- and 5-cm resolution, respectively. These data sets are composed by many large and fully annotated tiles, allowing an unbiased evaluation of models making use of spatial information. We do so by comparing two standard CNN architectures with the proposed one: standard patch classification, prediction of local label patches by employing only convolutions, and full patch labeling by employing deconvolutions. All the systems compare favorably or outperform a state-of-the-art baseline relying on superpixels and powerful appearance descriptors. The proposed full patch labeling CNN outperforms these models by a large margin, also showing a very appealing inference time.
Optimal Transport for Domain Adaptation
Courty, Nicolas ; Flamary, Remi ; Tuia, Devis ; Rakotomamonjy, Alain - \ 2017
IEEE Transactions on Pattern Analysis and Machine Intelligence 39 (2017)9. - ISSN 0162-8828 - p. 1853 - 1865.
classification - optimal transport - transfer learning - Unsupervised domain adaptation - visual adaptation
Domain adaptation is one of the most challenging tasks of modern data analytics. If the adaptation is done correctly, models built on a specific data representation become more robust when confronted to data depicting the same classes, but described by another observation system. Among the many strategies proposed, finding domain-invariant representations has shown excellent properties, in particular since it allows to train a unique classifier effective in all domains. In this paper, we propose a regularized unsupervised optimal transportation model to perform the alignment of the representations in the source and target domains. We learn a transportation plan matching both PDFs, which constrains labeled samples of the same class in the source domain to remain close during transport. This way, we exploit at the same time the labeled samples in the source and the distributions observed in both domains. Experiments on toy and challenging real visual adaptation examples show the interest of the method, that consistently outperforms state of the art approaches. In addition, numerical experiments show that our approach leads to better performances on domain invariant deep learning features and can be easily adapted to the semi-supervised case where few labeled samples are available in the target domain.
Spatial classification with fuzzy lattice reasoning
Mavridis, Constantinos ; Athanasiadis, I.N. - \ 2017
In: Proceedings of the 1st International Conference on Internet of Things and Machine Learning. - ACM - ISBN 9781450352437
Fuzzy Lattice Reasoning - classification - spatial data - spatial data mining - spatial classification - linear arrangement - data mining
This work extends the Fuzzy Lattice Reasoning (FLR) Classifier to manage spatial attributes, and spatial relationships. Specifically, we concentrate on spatial entities, as countries, cities, or states. Lattice Theory requires the elements of a Lattice to be partially ordered. To match such requirement, spatial entities are represented as a graph, whose number of nodes is equal to the amount of unique values of the spatial attribute elements. Then, the graph nodes are linearly arranged to formulate a partially ordered set; and thus be included in the Fuzzy Lattice classifier. The overall problem of incorporating spatial attributes in FLR was deduced to a Minimum Linear Arrangement problem. A corresponding open-source implementation in R has been made available on CRAN repository. The proposed method was evaluated using an open spatial dataset from the National Ambient Air Quality Standards (NAAQS). We investigated whether the addition of the spatial attribute contributed to any improvements in classification accuracy; and how linear arrangement alternatives may affect it. Experimental results showed that classification accuracy is above 85% in all cases, and the use of spatial attributes resulted to an increased accuracy of 92%. Alternative linear arrangements did not contribute significantly in improving classification accuracy in this case study.
Trajectories of agricultural change in southern Mali
Falconnier, G.N. - \ 2016
Wageningen University. Promotor(en): Ken Giller, co-promotor(en): Katrien Descheemaeker; T.A. van Mourik. - Wageningen : Wageningen University - ISBN 9789462577596 - 209
agriculture - agricultural development - farms - classification - self sufficiency - food - income - intensification - farming systems - intensive production - mali - landbouw - landbouwontwikkeling - landbouwbedrijven - classificatie - zelfvoorziening - voedsel - inkomen - intensivering - bedrijfssystemen - intensieve productie - mali
Key words: longitudinal study, farm typology, food self-sufficiency, income, legumes, ex-ante analysis, participatory research, scenario.
Smallholder agriculture in sub-Saharan Africa provides basis of rural livelihoods and food security, yet farmers have to cope with land constraints, variable rainfall and unstable institutional support. This study integrates a diversity of approaches (household typology and understanding of farm trajectories, on-farm trials, participatory ex-ante trade-off analysis) to design innovative farming systems to confront these challenges. We explored farm trajectories during two decades (1994 to 2010) in the Koutiala district in southern Mali, an area experiencing the land constraints that exert pressure in many other parts of sub-Saharan Africa. We classified farms into four types differing in land and labour productivity and food self-sufficiency status. During the past two decades, 17% of the farms stepped up to a farm type with greater productivity, while 70% of the farms remained in the same type, and only 13% of the farms experienced deteriorating farming conditions. Crop yields did not change significantly over time for any farm type and labour productivity decreased. Together with 132 farmers in the Koutiala district, we tested a range of options for sustainable intensification, including intensification of cereal (maize and sorghum) and legume (groundnut, soyabean and cowpea) sole crops and cereal-legume intercropping over three years and cropping seasons (2012-2014) through on-farm trials. Experiments were located across three soil types that farmers identified – namely black, sandy and gravelly soils. Enhanced agronomic performance was achieved when targeting legumes to a given soil type and/or place in the rotation: the biomass production of the cowpea fodder variety was doubled on black soils compared with gravelly soils and the additive maize/cowpea intercropping option after cotton or maize resulted in no maize grain penalty, and 1.38 t ha−1 more cowpea fodder production compared with sole maize. Farm systems were re-designed together with the farmers involved in the trials. A cyclical learning model combining the on-farm testing and participatory ex-ante analysis was used during four years (2012-2015). In the first cycle of 2012-2014, farmers were disappointed by the results of the ex-ante trade-off analysis, i.e marginal improvement in gross margin when replacing sorghum with soybean and food self-sufficiency trade-offs when intercropping maize with cowpea. In a second cycle in 2014-2015 the farm systems were re-designed using the niche-specific (soil type/previous crop combinations) information on yield and gross margin, which solved the concerns voiced by farmers during the first cycle. Farmers highlighted the saliency of the niches and the re-designed farm systems that increased farm gross margin by 9 to 29% (depending on farm type and options considered) without compromising food self-sufficiency. The involvement of farmers in the co-learning cycles allowed establishment of legitimate, credible and salient farm reconfiguration guidelines that could be scaled-out to other communities within the “old cotton basin”. Five medium-term contrasting socio-economic scenarios were built towards the year 2027, including hypothetical trends in policy interventions and change towards agricultural intensification. A simulation framework was built to account for household demographic dynamics and crop/livestock production variability. In the current situation, 45% of the 99 households of the study village were food self-sufficient and above the 1.25 US$ day-1 poverty line. Without change in farmer practices and additional policy intervention, only 16% of the farms would be both food self-sufficient and above the poverty line in 2027. In the case of diversification with legumes combined with intensification of livestock production and support to the milk sector, 27% of farms would be food self-sufficient and above the poverty line. Additional broader policy interventions to favour out-migration would be needed to lift 69% of the farms out of poverty. Other additional subsidies to favour yield gap narrowing of the main crops would lift 92% of the farm population out of poverty. Whilst sustainable intensification of farming clearly has a key role to play in ensuring food self-sufficiency, and is of great interest to local farmers, in the face of increasing population pressure other approaches are required to address rural poverty. These require strategic and multi-sectoral approaches that address employment within and beyond agriculture, in both rural and urban areas.
Een evaluatie van de maatlatten R6 en R7 voor de Kader Richtlijn Water
Griffioen, A.B. ; Vries, I. de - \ 2016
IMARES (Rapport / IMARES C087/15) - 28
rivieren - kaderrichtlijn water - waterbeheer - classificatie - waterkwaliteit - aquatische ecologie - monitoring - rivers - water framework directive - water management - classification - water quality - aquatic ecology - monitoring
De watertypes R6 en R7 in de Kader Richtlijn Water (KRW) classificering verschillen qua grootte van het waterlichaam en structuur. Het watertype R7 staat voor de grote rivieren met een hoofdstroom en nevengeulen. Rivieren als de Rijn, Waal en IJssel zijn hier voorbeelden van. Het watertype R6 staat voor langzaam stromende kleinere rivieren. In de praktijk kunnen beide riviertypen in elkaar overgaan en is het goed mogelijk dat het visbestand een grote overlap kent, maar volgens verschillende maatlatten worden beoordeeld. Dit onderzoek heeft tot doel het inzichtelijk maken van de indeling in beide watertypes. Ook wordt er gekeken naar de verschillen tussen de watertypen R6 en R7.
Identifying and naming plant-pathogenic fungi: past, present, and future
Crous, P.W. ; Hawksworth, D.L. ; Wingfield, M.J. - \ 2015
Annual Review of Phytopathology 53 (2015). - ISSN 0066-4286 - p. 247 - 267.
molecular systematics - polyphyletic nature - polyphasic approach - mycorrhizal fungi - species concepts - taxonomy - genus - classification - identification - chromatography
Scientific names are crucial in communicating knowledge about fungi. In plant pathology, they link information regarding the biology, host range, distribution, and potential risk. Our understanding of fungal biodiversity and fungal systematics has undergone an exponential leap, incorporating genomics, web-based systems, and DNA data for rapid identification to link species to metadata. The impact of our ability to recognize hitherto unknown organisms on plant pathology and trade is enormous and continues to grow. Major challenges for phytomycology are intertwined with the Genera of Fungi project, which adds DNA barcodes to known biodiversity and corrects the application of old, established names via epi- or neotypification. Implementing the one fungus–one name system and linking names to validated type specimens, cultures, and reference sequences will provide the foundation on which the future of plant pathology and the communication of names of plant pathogens will rest.
NSO-typering 2015; Typering van agrarische bedrijven in Nederland
Everdingen, W.H. van - \ 2015
Den Haag : LEI Wageningen UR (Nota / LEI 2015-084) - 36
landbouw bedrijven - bedrijven - bedrijfsvergelijking in de landbouw - bedrijfsvoering - opbrengsten - inkomsten uit het landbouwbedrijf - bedrijfsgrootte in de landbouw - bedrijfsgegevens - standaardisering - classificatie - agrarische economie - farming - businesses - farm comparisons - management - yields - farm income - farm size - farm accountancy data - standardization - classification - agricultural economics
In 2014 is voor de Nederlandse variant een nieuw kengetal geïntroduceerd: de Standaard Verdiencapaciteit (SVC) van bedrijven. Dat kengetal is ontwikkeld vanwege verschillen in marge tussen de sectoren. Met de SVC is de bedrijfsgrootte van bedrijven over bedrijfstypen heen meer gerelateerd aan arbeidsinzet en resultaat dan bij de Standaardopbrengst (SO) het geval is. De classificatie is gekoppeld aan de Landbouwtelling. De normen worden berekend voor de categorieën dieren en gewassen die in de Landbouwtelling worden uitgevraagd. Het doel van dit document is inzicht verschaffen in de achtergronden, rekenschema’s, indelingen en normen die bij de typering in gebruik zijn rond de Landbouwtelling van 2015. Achtereenvolgens komen in de volgende paragrafen de Standaardopbrengst (1), de NSO-typering (2), de Standaard Verdiencapaciteit (3) en het gebruik van de gegevens (4) aan bod.
Mapping Soil Properties of Africa at 250 m resolution: random forest significantly improve current predictions
Hengl, T. ; Heuvelink, G.B.M. ; Kempen, B. ; Leenaars, J.G.B. ; Walsh, M.G. ; Shepherd, K.D. ; Sila, A. ; Macmillan, R.A. ; Mendes de Jesus, J.S. ; Tamene, L. ; Tondoh, J.E. - \ 2015
PLoS ONE 10 (2015)6. - ISSN 1932-6203
continental-scale - maps - classification - surveillance - management - models - carbon - trees
80% of arable land in Africa has low soil fertility and suffers from physical soil problems. Additionally, significant amounts of nutrients are lost every year due to unsustainable soil management practices. This is partially the result of insufficient use of soil management knowledge. To help bridge the soil information gap in Africa, the Africa Soil Information Service (AfSIS) project was established in 2008. Over the period 2008–2014, the AfSIS project compiled two point data sets: the Africa Soil Profiles (legacy) database and the AfSIS Sentinel Site database. These data sets contain over 28 thousand sampling locations and represent the most comprehensive soil sample data sets of the African continent to date. Utilizing these point data sets in combination with a large number of covariates, we have generated a series of spatial predictions of soil properties relevant to the agricultural management—organic carbon, pH, sand, silt and clay fractions, bulk density, cation-exchange capacity, total nitrogen, exchangeable acidity, Al content and exchangeable bases (Ca, K, Mg, Na). We specifically investigate differences between two predictive approaches: random forests and linear regression. Results of 5-fold cross-validation demonstrate that the random forests algorithm consistently outperforms the linear regression algorithm, with average decreases of 15–75% in Root Mean Squared Error (RMSE) across soil properties and depths. Fitting and running random forests models takes an order of magnitude more time and the modelling success is sensitive to artifacts in the input data, but as long as quality-controlled point data are provided, an increase in soil mapping accuracy can be expected. Results also indicate that globally predicted soil classes (USDA Soil Taxonomy, especially Alfisols and Mollisols) help improve continental scale soil property mapping, and are among the most important predictors. This indicates a promising potential for transferring pedological knowledge from data rich countries to countries with limited soil data.
A Bayesian approach to combine Landsat and ALOS PALSAR time series for near real-time deforestation detection
Reiche, J. ; Bruin, S. de; Hoekman, D.H. ; Verbesselt, J. ; Herold, M. - \ 2015
Remote Sensing 7 (2015). - ISSN 2072-4292 - p. 4973 - 4996.
conditional-probability networks - remotely-sensed images - forest cover loss - tropical deforestation - brazilian amazon - accuracy assessment - classification - sar - disturbance - fusion
To address the need for timely information on newly deforested areas at medium resolution scale, we introduce a Bayesian approach to combine SAR and optical time series for near real-time deforestation detection. Once a new image of either of the input time series is available, the conditional probability of deforestation is computed using Bayesian updating, and deforestation events are indicated. Future observations are used to update the conditional probability of deforestation and, thus, to confirm or reject an indicated deforestation event. A proof of concept was demonstrated using Landsat NDVI and ALOS PALSAR time series acquired at an evergreen forest plantation in Fiji. We emulated a near real-time scenario and assessed the deforestation detection accuracies using three-monthly reference data covering the entire study site. Spatial and temporal accuracies for the fused Landsat-PALSAR case (overall accuracy = 87.4%; mean time lag of detected deforestation = 1.3 months) were consistently higher than those of the Landsat- and PALSAR-only cases. The improvement maintained even for increasing missing data in the Landsat time series.
Hypothesis: the sound of the individual metabolic phenotype? Acoustic detection of NMR experiments
Cacciatore, S. ; Saccenti, E. ; Piccioli, M. - \ 2015
OMICS - A Journal of Integrative Biology 19 (2015)3. - ISSN 1536-2310 - p. 147 - 156.
breast-cancer - personalized medicine - disease - profiles - models - health - time - classification - identification - metabonomics
We present here an innovative hypothesis and report preliminary evidence that the sound of NMR signals could provide an alternative to the current representation of the individual metabolic fingerprint and supply equally significant information. The NMR spectra of the urine samples provided by four healthy donors were converted into audio signals that were analyzed in two audio experiments by listeners with both musical and non-musical training. The listeners were first asked to cluster the audio signals of two donors on the basis of perceived similarity and then to classify unknown samples after having listened to a set of reference signals. In the clustering experiment, the probability of obtaining the same results by pure chance was 7.04% and 0.05% for non-musicians and musicians, respectively. In the classification experiment, musicians scored 84% accuracy which compared favorably with the 100% accuracy attained by sophisticated pattern recognition methods. The results were further validated and confirmed by analyzing the NMR metabolic profiles belonging to two other different donors. These findings support our hypothesis that the uniqueness of the metabolic phenotype is preserved even when reproduced as audio signal and warrants further consideration and testing in larger study samples
Remote sensing of epibenhtic shellfish using synthetic aperture radar satellite imagery
Nieuwhof, S. ; Herman, P.M.J. ; Dankers, N.M.J.A. ; Troost, K. ; Wal, D. van der - \ 2015
Remote Sensing 7 (2015)4. - ISSN 2072-4292 - p. 3710 - 3734.
bare soil surfaces - mussel beds - wadden sea - ecosystem engineers - intertidal flats - tidal flats - sar data - roughness - classification - moisture
On intertidal mudflats, reef-building shellfish, like the Pacific oyster and the blue mussel, provide a myriad of ecosystem services. Monitoring intertidal shellfish with high spatiotemporal resolution is important for fisheries, coastal management and ecosystem studies. Here, we explore the potential of X- (TerraSAR-X) and C-band (Radarsat-2) dual-polarized SAR data to map shellfish densities, species and coverage. We investigated two backscatter models (the integral equation model (IEM) and Oh’s model) for inversion possibilities. Surface roughness (vertical roughness RMSz and correlation length L) was measured of bare sediments and shellfish beds, which was then linked to shellfish density, presence and species. Oysters, mussels and bare sediments differed in RMSz, but because the backscatter saturates at relatively low RMSz values, it was not possible to retrieve shellfish density or species composition from X- and C-band SAR. Using a classification based on univariate and multivariate logistic regression of the field and SAR image data, we constructed maps of shellfish presence (Kappa statistics for calibration 0.56–0.74 for dual-polarized SAR), which were compared with independent field surveys of the contours of the beds (Kappa statistics of agreement 0.29–0.53 when using dual-polarized SAR). We conclude that spaceborne SAR allows one to monitor the contours of shellfish-beds (thus, distinguishing shellfish substrates from bare sediment and dispersed single shellfish), but not densities and species. Although spaceborne SAR cannot replace ground surveys entirely, it could very well offer a significant improvement in efficiency.
Discrimination of Polish unifloral honeys using overall PTR-MS and HPLC fingerprints combined with chemometrics
Kus, P.M. ; Ruth, S.M. van - \ 2015
Food Science and Technology = Lebensmittel-Wissenschaft und Technologie 62 (2015)1. - ISSN 0023-6438 - p. 69 - 75.
reaction-mass spectrometry - origin determination - botanical origin - electronic nose - floral markers - l. honey - volatile - classification - identification - flavonoids
A total of 62 honey samples of six floral origins (rapeseed, lime, heather, cornflower, buckwheat and black locust) were analysed by means of proton transfer reaction mass spectrometry (PTR-MS) and HPLC-DAD. The data were evaluated by principal component analysis and k-nearest neighbours classification in order to examine consistent differences in analytical fingerprints between various honeys allowing their discrimination. The study revealed, that both techniques were able to distinguish the floral origins, however the HPLC shows advantage over PTR-MS providing substantially better differentiation of all analysed honey types. Especially HPLC fingerprints recorded at 210 nm were most suitable for discrimination of botanical origin with the use of chemometric analysis. The obtained classification rates were: 100%, 93%, 100%, 83%, 100%, 100% (HPLC) and 69%, 67%, 78%, 67%, 100%, 88% (PTR-MS) for rapeseed, lime, heather, cornflower, buckwheat and black locust, respectively. Even if performance of PTR-MS in general was lower than HPLC, it might be useful for fast on-line screening of buckwheat honey.
Multi-model radiometric slope correction of SAR images of complex terrain using a two-stage semi-empirical approach
Hoekman, D.H. ; Reiche, J. - \ 2015
Remote Sensing of Environment 156 (2015). - ISSN 0034-4257 - p. 1 - 10.
radar imagery - topography - forest - classification - backscatter - validation
Practical approaches for the implementation of terrain type dependent radiometric slope correction for SAR data are introduced. Radiometric slope effects are modelled as the products of two models. The first is a simple physical model based on the assumption of a uniform opaque layer of isotropic scatterers, which is independent of terrain type, frequency and polarization. It accounts for the slope-induced variation in the number of scatterers per resolution cell. The second is a semi-empirical model, which accounts for the variation in scattering mechanisms, dependent on terrain type, frequency and polarization. PALSAR FBD (L-band, HH- and HV-polarization) data are used at two test sites in Brazil and Fiji. Results for the Brazilian area, which has slopes up to 25°, show that remaining slope effects for the multi-model case are much smaller than 0.1 dB, for all land cover types. This is much better than the best single-model approach where remaining slope effects can be very small for forests but be as large as 1.77 dB for woodland in HH-polarization. Results for the Fiji area, which has different vegetation types, are very similar. The potential large improvement, using this multi-model approach, in the accuracy of biomass estimation for transparent or open canopies is discussed. It is also shown that biomass change on slopes can be systematically under- or overestimated because of associated change in scattering mechanism.
Fusing Landsat and SAR time series to detect deforestation in the tropics
Reiche, J. ; Verbesselt, J. ; Hoekman, D.H. ; Herold, M. - \ 2015
Remote Sensing of Environment 156 (2015). - ISSN 0034-4257 - p. 276 - 293.
forest cover loss - alos palsar data - operational performance - accuracy assessment - multiscale texture - brazilian amazonia - missing data - jers-1 sar - satellite - classification
Fusion of optical and SAR time series imagery has the potential to improve forest monitoring in tropical regions, where cloud cover limits optical satellite time series observations. We present a novel pixel-based Multi-sensor Time-series correlation and Fusion approach (MulTiFuse) that exploits the full observation density of optical and SAR time series. We model the relationship of two overlapping univariate time series using an optimized weighted correlation. The resulting optimized regression model is used to predict and fuse two time series. Using the MulTiFuse approach we fused Landsat NDVI and ALOS PALSAR L-band backscatter time series. We subsequently used the fused time series in a multi-sensor change detection framework to detect deforestation between 01/2008 - 09/2010 at a managed forest plantation in the tropics (Pinus caribea; 2859 ha). 3-monthly reference data covering the entire study area was used to validate and assess spatial and temporal accuracy. We tested the impact of persistent cloud cover by increasing the per-pixel missing data percentage of the NDVI time series stepwise from ~ 53% (~ 6 observations/year) up to 95% (~ 0.5 observation/year) while fusing with a consistent PALSAR time series of ~ 2 observations/year. A significant linear correlation was found between the Landsat NDVI and ALOS PALSAR L-band SAR time series observables for logged forest. The multi-temporal filtered PALSAR HVHH backscatter ratio time series (HVHHmt) was most strongly correlated with the NDVI time series. While for Landsat-only the spatial and temporal accuracy of detected deforestation decreased significantly with increasing missing data, the accuracies for the fused NDVI-PALSAR case remained high and were observed to be above the NDVI- and PALSAR-only cases for all missing data percentages. For the fused NDVI-HVHHmt time series the overall accuracy was 95.5% with a 1.59 month mean time lag of detected changes. The MulTiFuse approach is robust and automated, and it provides the opportunity to use the upcoming data streams of free-of charge, medium resolution optical and SAR satellite imagery in a beneficial way for improved tropical forest monitoring.
Non-linear low-rank and sparse representation for hyperspectral image analysis
Morsier, Frank De; Tuia, Devis ; Borgeaucft, Maurice ; Gass, Volker ; Thiran, Jean Philippe - \ 2014
In: International Geoscience and Remote Sensing Symposium (IGARSS). - Institute of Electrical and Electronics Engineers Inc. (International Geoscience and Remote Sensing Symposium (IGARSS) ) - ISBN 9781479957750 - p. 4648 - 4651.
classification - kernel - low-rank - manifold clustering - sparse - unsupervised
In this paper, we tackle the problem of unsupervised classification of hyperspectral images. We propose a clustering method based on graphs representing the data structure, which is assumed to be an union of multiple manifolds. The method constraints the pixels to be expressed as a low-rank and sparse combination of the others in a reproducing kernel Hilbert spaces (RKHS). This captures the global (low-rank) and local (sparse) structures. Spectral clustering is applied on the graph to assign the pixels to the different manifolds. A large scale approach is proposed, in which the optimization is first performed on a subset of the data and then it is applied to the whole image using a non-linear collaborative representation respecting the manifolds structure. Experiments on two hyperspectral images show very good unsupervised classification results compared to competitive approaches.
Importance of bistatic SAR features from TanDEM-X for forest mapping and monitoring
Schlund, M. ; Poncet, F. von; Hoekman, D.H. ; Kuntz, S. ; Schmullius, C. - \ 2014
Remote Sensing of Environment 151 (2014)sp. issue. - ISSN 0034-4257 - p. 16 - 26.
land-cover - southeast-asia - feature-selection - polarimetric sar - tropical-forest - decision tree - alos palsar - rain-forest - sir-c - classification
Deforestation and forest degradation are one of the important sources for human induced carbon dioxide emissions and their rates are highest in tropical forests. For man-kind, it is of great importance to track land-use conversions like deforestation, e.g. for sustainable forest management and land use planning, for carbon balancing and to support the implementation of international initiatives like REDD + (Reducing Emissions from Deforestation and Degradation). SAR (synthetic aperture radar) sensors are suitable to reliably and frequently monitor tropical forests due to their weather independence. The TanDEM-X mission (which is mainly aimed to create a unique global high resolution digital elevation model) currently operates two X-band SAR satellites, acquiring interferometric SAR data for the Earth's entire land surface multiple times. The operational mission provides interferometric data as well as mono- and bistatic scattering coefficients. These datasets are homogenous, globally consistent and are acquired in high spatial resolution. Hence, they may offer a unique basic dataset which could be useful in land cover monitoring. Based on first datasets available from the TanDEM-X mission, the main goal of this research is to investigate the information content of TanDEM-X data for mapping forests and other land cover classes in a tropical peatland area. More specifically, the study explores the utility of bistatic features for distinguishing between open and closed forest canopies, which is of relevance in the context of deforestation and forest degradation monitoring. To assess the predominant information content of TanDEM-X data, the importance of information derived from the bistatic system is compared against the monostatic case, usually available from SAR systems. The usefulness of the TanDEM-X mission data, i.e. scattering coefficients, derived textural information and interferometric coherence is investigated via a feature selection process. The resulting optimal feature sets representing a monostatic and a bistatic SAR dataset were used in a subsequent classification to assess the added value of the bistatic TanDEM-X features in the separability of land cover classes. The results obtained indicated that especially the interferometric coherence significantly improved the separability of thematic classes compared to a dataset of monostatic acquisition. The bistatic coherence was mainly governed by volume decorrelation of forest canopy constituents and carries information about the canopy structure which is related to canopy cover. In contrast, the bistatic scattering coefficient had no significant contribution to class separability. The classification with coherence and textural information outperformed the classification with the monostatic scattering coefficient and texture by more than 10% and achieved an overall accuracy of 85%. These results indicate that TanDEM-X can serve as a valuable and consistent source for mapping and monitoring tropical forests.
Do Current European Policies Prevent Soil Threats and Support Soil Functions?
Glaesner, N. ; Helming, K. ; Vries, W. de - \ 2014
Sustainability 6 (2014)12. - ISSN 2071-1050 - p. 9538 - 9563.
ecosystem services - sustainable intensification - management - agriculture - protection - framework - quality - carbon - classification - conservation
There is currently no legislation at the European level that focuses exclusively on soil conservation. A cross-policy analysis was carried out to identify gaps and overlaps in existing EU legislation that is related to soil threats and functions. We found that three soil threats, namely compaction, salinization and soil sealing, were not addressed in any of the 19 legislative policies that were analyzed. Other soil threats, such as erosion, decline in organic matter, loss of biodiversity and contamination, were covered in existing legislation, but only a few directives provided targets for reducing the soil threats. Existing legislation addresses the reduction of the seven soil functions that were analyzed, but there are very few directives for improving soil functions. Because soil degradation is ongoing in Europe, it raises the question whether existing legislation is sufficient for maintaining soil resources. Addressing soil functions individually in various directives fails to account for the multifunctionality of soil. This paper suggests that a European Soil Framework Directive would increase the effectiveness of conserving soil functions in the EU.
Agroforestry solutions to address climate change and food security challenges in Africa
Mbow, C. ; Neufeldt, H. ; Noordwijk, M. van; Minang, P.A. ; Kowero, G. ; Luedeling, E. - \ 2014
Current Opinion in Environmental Sustainability 6 (2014). - ISSN 1877-3435 - p. 61 - 67.
sub-saharan africa - forest degradation - land degradation - climate-change - west-africa - agriculture - systems - intensification - classification - security
Trees inside and outside forests contribute to food security in Africa in the face of climate variability and change. They also provide environmental and social benefits as part of farming livelihoods. Varied ecological and socio-economic conditions have given rise to specific forms of agroforestry in different parts of Africa. Policies that institutionally segregate forest from agriculture miss opportunities for synergy at landscape scale. More explicit inclusion of agroforestry and the integration of agriculture and forestry agendas in global initiatives on climate change adaptation and mitigation can increase their effectiveness. We identify research gaps and overarching research questions for the contributions in this special issue that may help shape current opinion in environmental sustainability.