Object-based Detailed Vegetation Classification with Airborne High Spatial Resolution Remote Sensing Imagery

Abstract

In this paper, we evaluate the capability of the high spatial resolution airborne Digital Airborne Imaging System (DAIS) imagery for detailed vegetation classification at the alliance level with the aid of ancillary topographic data. Image objects as minimum classification units were generated through the Fractal Net Evolution Approach (FNEA) segmentation using eCognition software. For each object, 52 features were calculated including spectral features, textures, topographic features, and geometric features. After statistically ranking the importance of these features with the classification and regression tree algorithm (CART), the most effective features for classification were used to classify the vegetation. Due to the uneven sample size for each class, we chose a non-parametric (nearest neighbor) classifier. We built a hierarchical classification scheme and selected features for each of the broadest categories to carry out the detailed classification, which significantly improved the accuracy. Pixel-based maximum likelihood classification (MLC) with comparable features was used as a benchmark in evaluating our approach. The objectbased classification approach overcame the problem of saltand-pepper effects found in classification results from traditional pixel-based approaches. The method takes advantage of the rich amount of local spatial information present in the irregularly shaped objects in an image. This classification approach was successfully tested at Point Reyes National Seashore in Northern California to create a comprehensive vegetation inventory. Computer-assisted classification of high spatial resolution remotely sensed imagery has good potential to substitute or augment the present ground-based inventory of National Park lands.

Composition and spatial structure of dwarf shrub heath in Norway

“…The first step is the partition of the land surface into observation units. This can be done as a detailed point sample survey such as the BraunBlanquet approach (Westhoff & Van Der Maarel 1978), by field-based wall-to-wall mapping (KeelerWolf 2007), or by remote sensing (Yu et al 2006). The second step is the characterization of the vegetation found at each location.…”

Abordagens Para Classificação Do Estádio Sucessional Da Vegetação Do Parque Nacional De São Joaquim Empregando Imagens Landsat-8 E Rapideye

“…A menor acurácia dos experimentos conduzidos com as imagens RapidEye pode ser atribuída à maior variabilidade espectral no interior da mesma classe de cobertura da terra Este estudo mostrou que embora a resposta espectral tenha sido reconhecida como a característica mais importante na classificação da cobertura da terra, a informação textural inerente às imagens também pode prover elementos valiosos para melhorar a classificação (Li et al 2011, Lu et al 2012. A aplicação da informação textural foi apontada em vários estudos para melhor discriminação de classes de vegetação (Yu et al 2006, Sette e Maillard 2011, Roslani et al 2013, Gomes e Maillard 2015. Sette e Maillard (2011), ao classificarem os estádios sucessionais da vegetação de Floresta Ombrófila Densa em imagens do satélite FORMOSAT-2, no sul da Bahia, obtiveram acurácia de 60,5% ao utilizar somente as bandas do visível e 91% ao incluir atributos texturais.…”

Stratifying land use/land cover for spatial analysis of disease ecology and risk: an example using object-based classification techniques

“…In the first or segmentation step, spatial objects were formed. Objects are defined as groups of adjacent pixels treated as a single entity (Hay et al, 2001 size, shape and relationship to the overall image topology (Yu et al, 2006). Objects were segmented using the fractal net evolution approach (FNEA) as implemented in the eCognition software, v4.2 (Baatz et al, 2004).…”