A model-based performance test for forest classifiers on remote-sensing imagery

Abstract

Ambiguity between forest types on remote-sensing imagery is a major cause of errors found in accuracy assessments of forest inventorymaps. This paper presents a methodology, based on forest plot inventory, ground measurements and simulated imagery, for systematically quantifying these ambiguities in the sense of the minimum distance (MD), maximum likelihood (ML), and frequency-based (FB) classifiers. The method is tested with multi-spectral IKONOS images acquired on areas containing six major communities (oak, pine, fir, primary and secondary high tropical forests, and avocado plantation) of the National Forest Inventory (NFI) map in Mexico. A structural record of the canopy and optical measurements (leaf area index and soil reflectance) were performed on one plot of each class. Intra-class signal variation was modelled using the Discrete Anisotropic Radiative Transfer (DART) simulator of remote-sensing images. Atmospheric conditions were inferred from ground measurements on reference surfaces and leaf optical properties of each forest type were derived from the IKONOS forest signal. Next, all forest types were simulated, using a common environmental configuration, in order to quantify similarity among all forest types, according to MD, ML and FB classifiers. Classes were considered ambiguous when their dissimilarity was smaller than intra-class signal variation. DART proved useful in approximating the pixel value distribution and the ambiguity pattern measured on real forest imagery. In the case study, the oak forest and the secondary tropical forest were both distinguishable from all other classes using an MD classifier in a 25 m window size, whereas pine and primary tropical forests were ambiguous with three other classes using MD. By contrast, only two pairs of classes were found ambiguous for the ML classifier and only one for the FB classifier in that same window size. The avocado plantation was confounded with the primary tropical forest for all classifiers, presumably because the reflectance of both types of forest is governed by a deep canopy and a similar shadow area. We confronted the results of this study with the confusion matrix from the accuracy assessment of the NFI map. An asset of this model-basedmethod is its applicability to a variety of sensor types, eco-zones and class definitions

The Quality of Detailed Land Cover Maps in Highly Bio-Diverse Areas: Lessons Learned from the Mexican Experience

“…Third and last, ambiguity between classes on satellite imagery, related to the above situations, becomes more likely. In these conditions, the information on spectral separability could be a systematic tool to prioritise future cartographic efforts (Couturier et al, 2009b). Confronted with the three implications, we developed two methods based on recent theoretical advances made by the geo-science community.…”

Land use/land cover change and driving effects of water environment system in Dunhuang Basin, northwestern China

“…The series of conventional classifiers, such as parallelepipeds, minimum distance, and maximum likelihood models, are well-developed and have long been used for remote sensing applications (Couturier et al 2009;Gong et al 2011;Hagner and Reese 2007). Other machinelearning classifiers have also been effectively applied to LULC mapping, such as artificial neural networks, machine-learning decision trees, genetic algorithms, and support vector machines (Atkinson and Tatnall 1997;Benediktsson et al 1990;Fisher 2010;Gong et al 2011;Stavrakoudis et al 2010Stavrakoudis et al , 2011Volpi et al 2013).…”

Enhanced land use/cover classification using support vector machines and fuzzy k-means clustering algorithms

“…A series of conventional classification methods have been well developed and long used for remote sensing applications, which are parallelepiped, minimum distance, and maximum likelihood (ML) models. [3][4][5] Many other advanced classification techniques have also been introduced in the field of remote sensing classification, including artificial neural networks, machine-learning, decision trees, genetic algorithms, and support vector machines (SVM). [6][7][8][9][10] Machine learning algorithms are widely used classification algorithms during the past decades and some assessments of their relative performance compared to other classifiers have been conducted in the Amazon region.…”