Enhanced land use/cover classification of heterogeneous tropical landscapes using support vector machines and textural homogeneity

Paneque-Gálvez, Jaime; Mas, Jean-François; Moré, G.; Cristóbal, Jordi; Orta-Martínez, Martí; Luz, Ana Catarina; Guèze, Maximilien; Macía, Manuel J.; Reyes‐García, Victoria

doi:10.1016/j.jag.2012.10.007

Cited by 97 publications

(62 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Interestingly, using three times more features, the performance of RF and AdaBoost reached the one of SVM, whereas the latter remained the same; this indicated that SVM classifier was more sensitive to the Hughes phenomenon, resulting in decrease of classifier performance when the number of features is high compared with the number of training samples (Chi et al 2008). In other studies, SVM outperformed k-nearest neighbour (k-NN), binary Classification And Regression Tree (CART), and Maximum Likelihood Classifier (MLC) (Paneque-Gálvez et al 2013;Boyd et al 2006). The latter has outperformed Spectral Angle Mapper (SAM) and Spectral Information Divergence (SID) classifiers in a study by Forzieri et al (2013).…”

Section: Terrestrial Mappingmentioning

confidence: 91%

“…Landsat data, including the Multispectral Scanner (MSS), Thematic Mapper (TM), and Enhanced Thematic Mapper Plus (ETM+) sensors, are effective sources for terrestrial mapping, including habitat classification (Boyd et al 2006;Bock et al 2005), LC mapping of tropical areas (Paneque-Gálvez et al 2013), savannas (Sano et al 2010), grasslands (Price et al 2002), or forests (Jiang et al 2004;Wijedasa et al 2012), and change detection (Demir et al 2013;Berberoglu and Akin 2009). Other optical data include: the similar spatial resolution (i) Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) (Reiche et al 2012;Lucas et al 2011) and (ii) Linear Imaging Self Scanning Sensor 3 (LISS-III) (Lucas et al 2011); the higher resolution (iii) Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) (Vaglio Laurin et al 2013) and (iv) High Resolution Geometric (HRG) instrument (Lucas et al 2011); and the very high resolution (VHR) (v) QuickBird and (vi) WorldView-2 (Petrou et al 2014;Adamo et al 2014) sensors.…”

Section: Terrestrial Mappingmentioning

confidence: 99%

“…Mapping is performed through various classification approaches, including both knowledge (Lucas et al 2015;Pérez-Luque et al 2015;Petrou et al 2014;Adamo et al 2014) and data driven (Paneque-Gálvez et al 2013;Forzieri et al 2013;Chan et al 2012) ones. The former provide the flexibility to incorporate expert knowledge in the classifier and achieve rational classification results even in special cases or classes not included in the particular training site, although with the expense of labour and time consuming trial-and-error rule fine-tuning.…”

Section: Terrestrial Mappingmentioning

confidence: 99%

See 2 more Smart Citations

Remote sensing for biodiversity monitoring: a review of methods for biodiversity indicator extraction and assessment of progress towards international targets

2015

View full text Add to dashboard Cite

Recognizing the imperative need for biodiversity protection, the Convention on Biological Diversity (CBD) has recently established new targets towards 2020, the so-called Aichi targets, and updated proposed sets of indicators to quantitatively monitor the progress towards these targets. Remote sensing has been increasingly contributing to timely, accurate, and cost-effective assessment of biodiversity-related characteristics and functions during the last years. However, most relevant studies constitute individual research efforts, rarely related with the extraction of widely adopted CBD biodiversity indicators. Furthermore, systematic operational use of remote sensing data by managing authorities has still been limited. In this study, the Aichi targets and the related CBD indicators whose monitoring can be facilitated by remote sensing are identified. For each headline indicator a number of recent remote sensing approaches able for the extraction of related properties are reviewed. Methods cover a wide range of fields, including: habitat extent and condition monitoring; species distribution; pressures from unsustainable management, pollution and climate change; ecosystem service monitoring; and conservation status assessment of protected areas. The advantages and limitations of different remote sensing data and algorithms are discussed. Sorting of the methods based on their reported accuracies is attempted, when possible. The extensive literature survey aims at reviewing highly performing methods that can be used for large-area, effective, and timely biodiversity assessment, to encourage the more systematic use of remote sensing solutions in monitoring progress towards the Aichi targets, and to decrease the gaps between the remote sensing and management communities.

show abstract

Section: Terrestrial Mappingmentioning

confidence: 91%

Section: Terrestrial Mappingmentioning

confidence: 99%

Section: Terrestrial Mappingmentioning

confidence: 99%

See 1 more Smart Citation

Remote sensing for biodiversity monitoring: a review of methods for biodiversity indicator extraction and assessment of progress towards international targets

2015

View full text Add to dashboard Cite

show abstract

“…Texture analysis techniques have been applied to improve the classification accuracy [56] in different fields such as vegetation classification, land cover (e.g., [57][58][59][60][61][62]) and lithological mapping (e.g., [63]). The accuracy of the lithological map improves by using textures as additional layers, though the magnitude of the improvement is different from one rock type to another [63].…”

Section: Textural Indices For Lithological Classificationmentioning

confidence: 99%

Improving Lithological Mapping by SVM Classification of Spectral and Morphological Features: The Discovery of a New Chromite Body in the Mawat Ophiolite Complex (Kurdistan, NE Iraq)

2014

View full text Add to dashboard Cite

Abstract:The mineral ore potential of many mountainous regions of the world, like the Kurdistan region of Iraq, remains unexplored. For logistical and sometimes political reasons, these areas are difficult to map using traditional methods. We highlight the improvement in remote sensing geological mapping that arises from the integration of geomorphic features in classifications. The Mawat Ophiolite Complex (MOC) is located in the NE of Iraq and is known for its mineral deposits. The aims of this study are: (I) to refine the existing lithological map of the MOC; (II) to identify the best discriminatory datasets for lithological classification, including geomorphic features and textures; and (III) to identify potential locations with high concentrations of chromite. We performed a Support Vector Machine (SVM) classification method to allow the joint use of geomorphic features, textures and multispectral data of the Advanced Space-borne Thermal Emission and Reflection radiometer (ASTER) satellite. The updated map allowed the identification of a new mafic body and a substantial improvement of the geometry of the known lithological units. The use of geomorphic features allowed for the increase of the overall accuracy from 73% to 79.3%. In addition, we detected chromite occurrences within the ophiolite by applying Spectral Angle Mapping (SAM) technique. We identified two new locations having high concentrations of chromite and verified one of these promising areas and has coarsely crystalline chromite within dunite host rock. The chromium (Cr 2 O 3 ) concentration is ~8.46%. The SAM and SVM methods applied on ASTER satellite data show that these can be used as a powerful tool to explore ore deposits and to further improve lithological mapping in mountainous semi-arid regions.

show abstract

“…Additionally, because MAD-MEX is based on optical imagery and does not include specific techniques to attempt to detect forest degradation, it is clear that it cannot entirely meet the requirements of a fully-operational REDD+ MRV program. It is worth noting that there are already remote sensing techniques that can give estimations of degradation from optical data, such as Landsat [31,32], even if they are not able to account for all of the processes that cause anthropogenic forest degradation-, which may therefore be appropriate for the first wave of REDD+ MRV systems [33]. In countries like Mexico, the greater share of emissions from the land use, land use change and forestry sector may well stem from degradation, since so many forests are informally used by rural populations for grazing and for shifting cultivation, activities that result in lowered carbon levels, but not in long-run loss of forest cover.…”

Section: Implications Of Using Mad-mex For Activity Data Monitoring Imentioning

confidence: 99%

Comment on Gebhardt et al. MAD-MEX: Automatic Wall-to-Wall Land Cover Monitoring for the Mexican REDD-MRV Program Using All Landsat Data. Remote Sens. 2014, 6, 3923–3943

et al. 2016

Self Cite

View full text Add to dashboard Cite

Gebhardt et al. (2014) presented the Monitoring Activity Data for the Mexican REDD+ program (MAD-MEX), an automatic nation-wide land cover monitoring system for the Mexican REDD+ MRV. Though MAD-MEX represents a valuable first effort toward establishing a national reference emissions level for the implementation of REDD+ in Mexico, in this paper, we argue that this land cover system has important limitations that may prevent it from becoming operational for REDD+ MRV. Specifically, we show that (1) the accuracy assessment of MAD-MEX land cover maps is optimistically biased; (2) the ability of MAD-MEX to monitor land cover change, including deforestation and forest degradation; is poor and (3) the use of an entirely automatic classification approach, such as that followed by MAD-MEX, is highly problematic in the case of a large and heterogeneous country like Mexico. We discuss these limitations and call into question the ability of a land cover monitoring system, such as MAD-MEX, both to elaborate a national reference emissions level and to monitor future forest cover change, as part of a REDD+ MRV system. We provide some insights with the aim of improving the development of nation-wide land cover monitoring systems in Mexico and elsewhere.

show abstract

Enhanced land use/cover classification of heterogeneous tropical landscapes using support vector machines and textural homogeneity

Cited by 97 publications

References 41 publications

Remote sensing for biodiversity monitoring: a review of methods for biodiversity indicator extraction and assessment of progress towards international targets

Remote sensing for biodiversity monitoring: a review of methods for biodiversity indicator extraction and assessment of progress towards international targets

Improving Lithological Mapping by SVM Classification of Spectral and Morphological Features: The Discovery of a New Chromite Body in the Mawat Ophiolite Complex (Kurdistan, NE Iraq)

Comment on Gebhardt et al. MAD-MEX: Automatic Wall-to-Wall Land Cover Monitoring for the Mexican REDD-MRV Program Using All Landsat Data. Remote Sens. 2014, 6, 3923–3943

Contact Info

Product

Resources

About