Comparison between artificial neural networks and maximum likelihood classification in digital soil mapping

Soil surveys are the main source of spatial information on soils and have a range of different applications, mainly in agriculture. The continuity of this activity has however been severely compromised, mainly due to a lack of governmental funding. The purpose of this study was to evaluate the feasibility of two different classifiers (artificial neural networks and a maximum likelihood algorithm) in the prediction of soil classes in the northwest of the state of Rio de Janeiro. Terrain attributes such as elevation, slope, aspect, plan curvature and compound topographic index (CTI) and indices of clay minerals, iron oxide and Normalized Difference Vegetation Index (NDVI), derived from Landsat 7 ETM+ sensor imagery, were used as discriminating variables. The two classifiers were trained and validated for each soil class using 300 and 150 samples respectively, representing the characteristics of these classes in terms of the discriminating variables. According to the statistical tests, the accuracy of the classifier based on artificial neural networks (ANNs) was greater than of the classic Maximum Likelihood Classifier (MLC). Comparing the results with 126 points of reference showed that the resulting ANN map (73.81 %) was superior to the MLC map (57.94 %). The main errors when using the two classifiers were caused by: a) the geological heterogeneity of the area coupled with problems related to the geological map; b) the depth of lithic contact and/or rock exposure, and c) problems with the environmental correlation model used due to the polygenetic nature of the soils. This study confirms that the use of terrain attributes together with remote sensing data by an ANN approach can be a tool to facilitate soil mapping in Brazil, primarily due to the availability of low-cost remote sensing data and the ease by which terrain attributes can be obtained.

Building predictive models of soil particle-size distribution

Is it possible to build predictive models (PMs) of soil particle-size distribution (psd) in a region with complex geology and a young and unstable land-surface? The main objective of this study was to answer this question. A set of 339 soil samples from a small slope catchment in Southern Brazil was used to build PMs of psd in the surface soil layer. Multiple linear regression models were constructed using terrain attributes (elevation, slope, catchment area, convergence index, and topographic wetness index). The PMs explained more than half of the data variance. This performance is similar to (or even better than) that of the conventional soil mapping approach. For some size fractions, the PM performance can reach 70 %. Largest uncertainties were observed in geologically more complex areas. Therefore, significant improvements in the predictions can only be achieved if accurate geological data is made available. Meanwhile, PMs built on terrain attributes are efficient in predicting the particle-size distribution (psd) of soils in regions of complex geology.

Cinema Roads to the Platonic Image of the Cave

The aim of this article is to prove the real possibility of travelling intellectually to the Platonic image of the cave from different films. In this sense, one can speak of explicit references as in The Conformist by B. Bertolucci or in Shadowlands by R. Attenborough -if one bears in mind the Chronicles of Narnia by C. S. Lewis- or The Picture of Dorian Gray ¿if one bears in mind the well-known O. Wilde¿s novel-, but, on other occasions, although the Platonic influence cannot be proved, for instance in The Truman Show, A Room with a View or Brideshead Revisited, one can perfectly think of these films in order to guide the contemporary audiences to that Platonic image, since Plato himself affirms that it deals with an image which can be easily applied and, in first place, to his idealistic philosophy.

Improving Satellite Image Based Forest Inventory by Using A Priori Site Quality Information

Summary

The Cave by José Saramago: Platonic image versus metaphysics

The cave by José Saramago has as a certain reference the image of the cave of book VII of Plato's Republic and, however, Saramago is not an idealistic or metaphysical writer. This article, taking advantage of the applicability with which Plato endowed his image, defends the urge to be open to the messages sent by the earth, by matter, the urge not to become prisoners in the golden caves of the Western society and, finally, the urge to find our freedom in Nature, phýsis, and not far or beyond, metá, it.

The "image" of the cave and the constant temptation to correct Plato: Benjamin Jowettt as an example

Translations of the first chapters of Book VII of Plato's Republic, in which he introduces the well-known image of the cave, eikón, reveals an astonishing and intriguing variety of interpretations of this image: "allegory", "myth", "fable", "parable", "simile" and "comparison", to cite but a few. Taking as an example the work by Benjamin Jowett, the Victorian translator of Plato, remarkable for its textual accuracy and by means of a close analysis of the terms related to the image, this paper insists on the need to neither interpret nor correct the great ideal philosopher, in this case revealing some evident contradictions that arise when this advice is not followed and pointing out the occasional use of terms extraneous to the Platonic lexicon such as "allegory".

Influence of field slope and coffee plantation age on the physical properties of a red-yellow Latosol

In modern agriculture, several factors cause changes in the soil physical properties. The time of establishment of a crop (plantation age) and the slope are examples of factors that moderate the impact of mechanized operations on the soil structure. The objective of this study was to analyze the effect of machinery traffic on the physical properties of a Red-Yellow Latosol under coffee plantations with different ages (2, 7, 18, and 33 years) and slope positions (3, 9 and 15 %). Samples were collected from three positions between coffee rows (lower wheel track, inter-row and upper wheel track) and at two depths (surface layer and sub-surface). Changes in the total porosity, macroporosity, microporosity, organic matter, bulk density, and aggregate stability were investigated. Our results showed that the slope influenced the organic matter content, microporosity and aggregate stability. The soil samples under the inter-row were minimally damaged in their structure, compared to those from under the lower and upper wheel track, while the structure was better preserved under the lower than the upper track. The time since the establishment of the crop, i.e., the plantation age, was the main factor determining the extent of structural degradation in the coffee plantation.

Public-domain software for root image analysis

In the search for high efficiency in root studies, computational systems have been developed to analyze digital images. ImageJ and Safira are public-domain systems that may be used for image analysis of washed roots. However, differences in root properties measured using ImageJ and Safira are supposed. This study compared values of root length and surface area obtained with public-domain systems with values obtained by a reference method. Root samples were collected in a banana plantation in an area of a shallower Typic Carbonatic Haplic Cambisol (CXk), and an area of a deeper Typic Haplic Ta Eutrophic Cambisol (CXve), at six depths in five replications. Root images were digitized and the systems ImageJ and Safira used to determine root length and surface area. The line-intersect method modified by Tennant was used as reference; values of root length and surface area measured with the different systems were analyzed by Pearson's correlation coefficient and compared by the confidence interval and t-test. Both systems ImageJ and Safira had positive correlation coefficients with the reference method for root length and surface area data in CXk and CXve. The correlation coefficient ranged from 0.54 to 0.80, with lowest value observed for ImageJ in the measurement of surface area of roots sampled in CXve. The IC (95 %) revealed that root length measurements with Safira did not differ from that with the reference method in CXk (-77.3 to 244.0 mm). Regarding surface area measurements, Safira did not differ from the reference method for samples collected in CXk (-530.6 to 565.8 mm²) as well as in CXve (-4231 to 612.1 mm²). However, measurements with ImageJ were different from those obtained by the reference method, underestimating length and surface area in samples collected in CXk and CXve. Both ImageJ and Safira allow an identification of increases or decreases in root length and surface area. However, Safira results for root length and surface area are closer to the results obtained with the reference method.

Artificial neural networks applied for soil class prediction in mountainous landscape of the Serra do Mar¹

Soil information is needed for managing the agricultural environment. The aim of this study was to apply artificial neural networks (ANNs) for the prediction of soil classes using orbital remote sensing products, terrain attributes derived from a digital elevation model and local geology information as data sources. This approach to digital soil mapping was evaluated in an area with a high degree of lithologic diversity in the Serra do Mar. The neural network simulator used in this study was JavaNNS and the backpropagation learning algorithm. For soil class prediction, different combinations of the selected discriminant variables were tested: elevation, declivity, aspect, curvature, curvature plan, curvature profile, topographic index, solar radiation, LS topographic factor, local geology information, and clay mineral indices, iron oxides and the normalized difference vegetation index (NDVI) derived from an image of a Landsat-7 Enhanced Thematic Mapper Plus (ETM+) sensor. With the tested sets, best results were obtained when all discriminant variables were associated with geological information (overall accuracy 93.2 - 95.6 %, Kappa index 0.924 - 0.951, for set 13). Excluding the variable profile curvature (set 12), overall accuracy ranged from 93.9 to 95.4 % and the Kappa index from 0.932 to 0.948. The maps based on the neural network classifier were consistent and similar to conventional soil maps drawn for the study area, although with more spatial details. The results show the potential of ANNs for soil class prediction in mountainous areas with lithological diversity.

Advanced Kernel Methods For Remote Sensing Image Classification

Résumé Suite aux recentes avancées technologiques, les archives d'images digitales ont connu une croissance qualitative et quantitative sans précédent. Malgré les énormes possibilités qu'elles offrent, ces avancées posent de nouvelles questions quant au traitement des masses de données saisies. Cette question est à la base de cette Thèse: les problèmes de traitement d'information digitale à très haute résolution spatiale et/ou spectrale y sont considérés en recourant à des approches d'apprentissage statistique, les méthodes à noyau. Cette Thèse étudie des problèmes de classification d'images, c'est à dire de catégorisation de pixels en un nombre réduit de classes refletant les propriétés spectrales et contextuelles des objets qu'elles représentent. L'accent est mis sur l'efficience des algorithmes, ainsi que sur leur simplicité, de manière à augmenter leur potentiel d'implementation pour les utilisateurs. De plus, le défi de cette Thèse est de rester proche des problèmes concrets des utilisateurs d'images satellite sans pour autant perdre de vue l'intéret des méthodes proposées pour le milieu du machine learning dont elles sont issues. En ce sens, ce travail joue la carte de la transdisciplinarité en maintenant un lien fort entre les deux sciences dans tous les développements proposés. Quatre modèles sont proposés: le premier répond au problème de la haute dimensionalité et de la redondance des données par un modèle optimisant les performances en classification en s'adaptant aux particularités de l'image. Ceci est rendu possible par un système de ranking des variables (les bandes) qui est optimisé en même temps que le modèle de base: ce faisant, seules les variables importantes pour résoudre le problème sont utilisées par le classifieur. Le manque d'information étiquétée et l'incertitude quant à sa pertinence pour le problème sont à la source des deux modèles suivants, basés respectivement sur l'apprentissage actif et les méthodes semi-supervisées: le premier permet d'améliorer la qualité d'un ensemble d'entraînement par interaction directe entre l'utilisateur et la machine, alors que le deuxième utilise les pixels non étiquetés pour améliorer la description des données disponibles et la robustesse du modèle. Enfin, le dernier modèle proposé considère la question plus théorique de la structure entre les outputs: l'intègration de cette source d'information, jusqu'à présent jamais considérée en télédétection, ouvre des nouveaux défis de recherche. Advanced kernel methods for remote sensing image classification Devis Tuia Institut de Géomatique et d'Analyse du Risque September 2009 Abstract The technical developments in recent years have brought the quantity and quality of digital information to an unprecedented level, as enormous archives of satellite images are available to the users. However, even if these advances open more and more possibilities in the use of digital imagery, they also rise several problems of storage and treatment. The latter is considered in this Thesis: the processing of very high spatial and spectral resolution images is treated with approaches based on data-driven algorithms relying on kernel methods. In particular, the problem of image classification, i.e. the categorization of the image's pixels into a reduced number of classes reflecting spectral and contextual properties, is studied through the different models presented. The accent is put on algorithmic efficiency and the simplicity of the approaches proposed, to avoid too complex models that would not be used by users. The major challenge of the Thesis is to remain close to concrete remote sensing problems, without losing the methodological interest from the machine learning viewpoint: in this sense, this work aims at building a bridge between the machine learning and remote sensing communities and all the models proposed have been developed keeping in mind the need for such a synergy. Four models are proposed: first, an adaptive model learning the relevant image features has been proposed to solve the problem of high dimensionality and collinearity of the image features. This model provides automatically an accurate classifier and a ranking of the relevance of the single features. The scarcity and unreliability of labeled. information were the common root of the second and third models proposed: when confronted to such problems, the user can either construct the labeled set iteratively by direct interaction with the machine or use the unlabeled data to increase robustness and quality of the description of data. Both solutions have been explored resulting into two methodological contributions, based respectively on active learning and semisupervised learning. Finally, the more theoretical issue of structured outputs has been considered in the last model, which, by integrating outputs similarity into a model, opens new challenges and opportunities for remote sensing image processing.