What’s wrong with pixels? Some recent developments interfacing remote sensing and GIS

By Thomas Blaschke, J. Strobl. 2001

734 citations recorded by the source corpus.

How researchers see this paper

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

In Remote Sensing

Citations in the corpus

A multi-scale segmentation/object relationship modelling methodology for landscape analysis (2003, methodology): “Especially within the last two years, many new segmentation algorithms and applications have been tested in geoscience applications, but few of them lead to qualitatively convincing results while still being robust and operational (Blaschke and Strobl, 2001).”
Linking earth observation and taxonomic, structural and functional biodiversity: local to ecosystem perspectives (2016, methodology): “For this reason, the pixel-based approach has often been subject to criticism (Fisher, 1997; Blaschke and Strobl, 2001; Blaschke, 2010).”
Forest Type Mapping using Object-specific Texture Measures from Multispectral Ikonos Imagery: Segmentation Quality and Image Classification Issues (2009, background): “…as an important research area since the late 1990s, partly in the hope that this approach will emulate the human interpreters’ ability to identify and delineate features of interest (Blaschke and Strobl, 2001; Schiewe et al., 2001; Blaschke, 2003; Benz et al., 2004; Meinel and Neubert, 2004).”
The role of the spatial dimension within the framework of sustainable landscapes and natural capital (2006, background): “f i v v c s a a o o a i t s h information extraction from raw data exist (Blaschke and Strobl, 2001).”
Object‐based land‐cover classification for the Phoenix metropolitan area: optimization vs. transportability (2008, methodology): “…has been suggested that classifying remotely sensed images in pixelwise fashion (using only the first three pixel attributes) is a special case of the super-set of object-based classification methodologies which utilize all four (de Kok et al. 2000, Blaschke and Strobl 2001, Blaschke et al. 2004).”
Remote Sensing Image Scene Classification Meets Deep Learning: Challenges, Methods, Benchmarks, and Opportunities (2020, background): “In 2001, Blaschke and Strobl [29] questioned the dominance of per-pixel research paradigm and concluded that analyzing remote sensing images at object level is more efﬁcient than per-pixel analysis.”
Object-based contextual image classification built on image segmentation (2003, methodology): “While this was claimed earlier as hypotheses [14, 7] it could recently empirically verified by [16].”
Change detection from remotely sensed images: From pixel-based to object-based approaches (2013, background): “Another important limitation of the traditional pixel-based CD approaches is the difficulty of modeling the contextual information (Blaschke and Strobl, 2001; Johansen et al., 2010).”
Land-cover classification with high-resolution remote sensing images using transferable deep models (2018, methodology): “These methods can interpret RS images using the spectral features of individual pixels [10–12], but their performance is often heavily aﬀected by intra-class spectral variations and noises [50–52].”
Object based image analysis for remote sensing (2010, background): “…published around the year 2000, before the use of the terms ‘object-oriented image analysis’ or ‘objectbased image analysis’ (Blaschke et al., 2000; Blaschke and Strobl, 2001; Blaschke and Hay, 2001) therewere some influential articles that can be mentioned specifically as having been…”
Unsupervised image segmentation evaluation and refinement using a multi-scale approach (2011, positive): “However, traditional pixel-based image analysis methods, which extract information using statistics of single pixels, are often not well suited for this high-resolution imagery (Blaschke and Strobl, 2001; Benz et al., 2004).”
Preliminary evaluation of eCognition object-based software for cut block delineation and feature extraction (2003, methodology): “In 2000, a prototype object-based processing software program was available from Definiens Imaging GmbH that claimed to be user-friendly multiscaled, and fully functional (Blaschke and Strobl, 2001).”
Automated classification of landform elements using object-based image analysis (2006, methodology): “Advanced pixel-based processes such as texture measurements, linear mixture modeling, fuzzy sets and neural network classifiers were invented to enhance per-pixel image analysis (Blaschke and Strobl, 2001).”
AID: A Benchmark Data Set for Performance Evaluation of Aerial Scene Classification (2016, background): “Specifically, in 2001, Blaschke and Strobl [44] have raised the question “What’s wrong with pixels?” and argued that it is more efficient to analyze aerial images at object-level, where the “objects” refer to local regions of pixels sharing spectral or texture homogeneities, e.g. superpixels [45].”
ESP: a tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data (2010, background): “Object-based image analysis (OBIA) arose through the realization that image-objects hold more real-world value than pixels alone (Fisher 1997, Blaschke and Strobl 2001, Smith et al .”