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

By Qian Yu, Peng Gong, N. Clinton, G. Biging, Maggi Kelly, D. Schirokauer. 2006

814 citations recorded by the source corpus.

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.

Subfields: Ecological Modeling, Ecology, Media Technology

View Paper: https://www.semanticscholar.org/paper/c00a273c4b0c487e0068f28ccece47eacffbb210

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecological Modeling

In Ecology

In Media Technology

Citations in the corpus

Implementation of machine-learning classification in remote sensing: an applied review (2018, methodology): “For example, k -NN only requires setting the k parameter (Yu et al. 2006; Mallinis et al. 2008), and the Adaboost implementation of boosted DT only require the number of trees in the ensemble (Freund and Schapire 1996).”
Impact of sudden oak death on tree mortality in the Big Sur ecoregion of California (2008, background): “Object-based classification is especially useful for generalizing spectral information in high resolution imagery while maintaining its spatial detail (Luscier et al. 2006; Yu et al. 2006).”
Observing Changing Ecological Diversity in the Anthropocene (2013, background): “…forest (S) AVIRIS 20.0 Martin et al. (1998) Temperate needleleaf forest (S) DAIS 1.0 Gong et al. (1997) Temperate shrubland-grassland (F) DAIS 1.0 Yu et al. (2006) Temperate wetlands (F) CASI, other 1.0-3.0 Belluco et al. (2006) Temperate wetlands (S) HyMap 3.0 Lucas and Carter (2008) Tropical…”
An SVM Ensemble Approach Combining Spectral, Structural, and Semantic Features for the Classification of High-Resolution Remotely Sensed Imagery (2013, background): “However, it should be recognized that although various spatial features are currently available for high-resolution image processing, such as morphological features [6]–[10], [12], structural feature set [29], PSI [13], wavelet-based texture [30], object-based features [32], and GLCM [4], [33], it i”
Object-based urban vegetation mapping with high-resolution aerial photography as a single data source (2013, methodology): “It is difficult to use medium spatial resolution satellite imagery, such as Landsat Thematic Mapper (TM) and Système Pour l’Observation de la Terre (SPOT), to delineate detailed vegetation types or patterns in urban landscapes (Yu et al. 2006).”
Texture and Scale in Object-Based Analysis of Subdecimeter Resolution Unmanned Aerial Vehicle (UAV) Imagery (2009, methodology): “spectral information, or a low signal-to-noise ratio [15], [16].”
UAV Remote Sensing for Urban Vegetation Mapping Using Random Forest and Texture Analysis (2015, background): “However, single pixels no longer capture the characteristics of classification targets in HR images [17], and a “salt and pepper” effect is always obviously shown in the classification results.”
Geographic Object-Based Image Analysis – Towards a new paradigm (2014, result): “…articles that have shown that object-based classiﬁcation results (especially of H-resolution imagery) are consistently better than those based on traditional pixel-based approaches (e.g., Shackelford and Davis, 2003; Yu et al, 2006; Blaschke, 2010; Myint et al., 2011; Whiteside et al., 2011).”
Land-Use Scene Classification Using a Concentric Circle-Structured Multiscale Bag-of-Visual-Words Model (2014, background): “, color/spectrum, texture, and shape)-based classification methods [8]–[11], which are good at dealing with land-cover classification problems, can no longer achieve satisfactory results in the research of land-use scene classification.”
Measuring the Effectiveness of Various Features for Thematic Information Extraction From Very High Resolution Remote Sensing Imagery (2015, background): “are always characterized by different features based on plant biology, environmental factors, and so on [13].”
A Survey on Object Detection in Optical Remote Sensing Images (2016, neutral): “…et al., 2011; Phinn et al., 2012; Tzotsos et al., 2011; Walker and Blaschke, 2008; Walker and Briggs, 2007; Weng, 2009, 2011; Xie et al., 2008; Yu et al., 2006; Zhou, 2013; Zhou et al., 2009; Zhou and Troy, 2008); landslide mapping (Feizizadeh et al., 2014; Li et al., 2015b; Martha et al.,…”
Comparing object-based and pixel-based classifications for mapping savannas (2011, background): “Myint et al. (2011) found that including principal component images and a Normalized Difference Vegetation Index within an object-based rule-set classification produced significantly higher accuracy than MLC. Yu et al. (2006) incorporated ancillary and derivative data (IHS transformations, elevation”
Impact of training and validation sample selection on classification accuracy and accuracy assessment when using reference polygons in object-based classification (2013, background): “Thus, object-based classifiers efficiently avoid the ‘salt and pepper’ effect that frequently occurs in pixel-based classification (Yu et al. 2006; Blaschke 2010; Pu, Landry, and Yu 2011), especially for high-spatial resolution imagery.”