Comparing object-based and pixel-based classifications for mapping savannas

Whiteside, Timothy G.; Boggs, Guy S.; Maier, Stefan W.

doi:10.1016/j.jag.2011.06.008

Cited by 269 publications

(186 citation statements)

References 17 publications

Supporting

Mentioning

171

Contrasting

Unclassified

Order By: Relevance

“…For the classification analysis of images with very high spatial resolution, object-based approaches are superior over pixel-based approaches when the pixel size is significantly smaller than the average size of the objects of interest [37,38]. Immitzer et al also found that by classifying objects instead of pixels, the user accuracies could increase significantly for most tree species in a forest study of a temperate zone [2], and the positive impact was higher for conifers than broadleaved trees.…”

Section: Discussionmentioning

confidence: 99%

Interpretation of Forest Resources at the Individual Tree Level at Purple Mountain, Nanjing City, China, Using WorldView-2 Imagery by Combining GPS, RS and GIS Technologies

et al. 2013

View full text Add to dashboard Cite

This study attempted to measure forest resources at the individual tree level using high-resolution images by combining GPS, RS, and Geographic Information System (GIS) technologies. The images were acquired by the WorldView-2 satellite with a resolution of 0.5 m in the panchromatic band and 2.0 m in the multispectral bands. Field data of 90 plots were used to verify the interpreted accuracy. The tops of trees in three groups, namely ≥10 cm, ≥15 cm, and ≥20 cm DBH (diameter at breast height), were extracted by the individual tree crown (ITC) approach using filters with moving windows of 3 × 3 pixels, 5 × 5 pixels and 7 × 7 pixels, respectively. In the study area, there were 1,203,970 trees of DBH over 10 cm, and the interpreted accuracy was 73.68 ± 15.14% averaged over the 90 plots. The numbers of the trees that were ≥15 cm and ≥20 cm DBH were 727,887 and 548,919, with an average accuracy of 68.74 ± 17.21% and 71.92 ± 18.03%, respectively. The pixel-based classification showed that the classified accuracies of the 16 classes obtained using the eight multispectral bands were higher than those obtained using only the four standard bands. The increments ranged from 0.1% for the water class to 17.0% for Metasequoia glyptostroboides, with an average value of 4.8% for the 16 classes. In addition, to overcome the "mixed pixels" problem, a crown-based supervised classification, which can improve the classified accuracy of both dominant OPEN ACCESSRemote Sens. 2014, 6 88 species and smaller classes, was used for generating a thematic map of tree species. The improvements of the crown-to pixel-based classification ranged from −1.6% for the open forest class to 34.3% for Metasequoia glyptostroboides, with an average value of 20.3% for the 10 classes. All tree tops were then annotated with the species attributes from the map, and a tree count of different species indicated that the forest of Purple Mountain is mainly dominated by Quercus acutissima, Liquidambar formosana and Pinus massoniana. The findings from this study lead to the recommendation of using the crown-based instead of the pixel-based classification approach in classifying mixed forests.

show abstract

Section: Discussionmentioning

confidence: 99%

Interpretation of Forest Resources at the Individual Tree Level at Purple Mountain, Nanjing City, China, Using WorldView-2 Imagery by Combining GPS, RS and GIS Technologies

et al. 2013

View full text Add to dashboard Cite

show abstract

“…Next, the objects are classified using spectral as well as spatial information such as texture, shape and context features to more clearly distinguish spectrally similar land cover types. Most comparative studies that have compared OBIA and pixel classification approaches reported that OBIA provides the most accurate results (e.g., [19][20][21][22][23]). Recent advances in OBIA have revolutionized the processing of high to very high spatial resolution remote sensing data by providing effective computer-assisted classification techniques whose results are close to the quality of manual photo-interpretation, while also being much faster, cheaper and reproducible over large areas (e.g., [6,7]).…”

Section: Mapping Artificial Areas From Obiamentioning

confidence: 99%

“…Considering the size of the study area (and mapping regions), collecting field data for the control sample would normally be extremely labor-intensive and time-consuming. As suggested by [22] and [36], selected control objects were photo-interpreted using the image with the highest spatial resolution as control data. To ensure photo interpretation was objective, the classified map was not viewed during the process and the manual photo interpretation was made by an experienced photo-interpreter who was not involved in developing the method.…”

Section: Map Validationmentioning

confidence: 99%

An Object-Based Image Analysis Method for Monitoring Land Conversion by Artificial Sprawl Use of RapidEye and IRS Data

Dupuy

Barbe²,

Balestrat³

2012

Remote Sensing

View full text Add to dashboard Cite

Abstract:In France, in the peri-urban context, urban sprawl dynamics are particularly strong with huge population growth as well as a land crisis. The increase and spreading of built-up areas from the city centre towards the periphery takes place to the detriment of natural and agricultural spaces. The conversion of land with agricultural potential is all the more worrying as it is usually irreversible. The French Ministry of Agriculture therefore needs reliable and repeatable spatial-temporal methods to locate and quantify loss of land at both local and national scales. The main objective of this study was to design a repeatable method to monitor land conversion characterized by artificial sprawl: (i) We used an object-based image analysis to extract artificial areas from satellite images; (ii) We built an artificial patch that consists of aggregating all the peripheral areas that characterize artificial areas. The "artificialized" patch concept is an innovative extension of the urban patch concept, but differs in the nature of its components and in the continuity distance applied; (iii) The diachronic analysis of artificial patch maps enables characterization of artificial sprawl. The method was applied at the scale of four departments (similar to provinces) along the coast of Languedoc-Roussillon, in the South of France, based on two satellite datasets, one acquired in 1996-1997 (Indian Remote Sensing) and the other in 2009 (RapidEye). In the four departments, we measured an increase in artificial areas of from 113,000 ha in 1997 to 133,000 ha in 2009, i.e., an 18% increase in 12 years. The package comes in the form of a 1/15,000 valid cartography, usable at the scale of a commune (the smallest territorial division used for administrative purposes in France) that can be adapted to departmental and regional scales. The method is reproducible in OPEN ACCESSRemote Sens. 2012, 4 405 homogenous spatial-temporal terms, so that it could be used periodically to assess changes in land conversion rates in France as a whole.

show abstract

“…The approach allows use of multiple image elements, parameters and scales such as texture, shape and context, as opposed to pixel-based classification that solely relies on the pixel value. Overall, OBIA has been proven to produce more accurate classification results compared to pixel-based approaches using medium to high resolution imagery, producing improvements in classification accuracy ranging from 9-23% [18][19][20].…”

Section: Introductionmentioning

confidence: 99%

Mapping Net Stocked Plantation Area for Small-Scale Forests in New Zealand Using Integrated RapidEye and LiDAR Sensors

Xu¹,

Morgenroth²,

Manley³

2017

Forests

View full text Add to dashboard Cite

Abstract:In New Zealand, approximately 70% of plantation forests are large-scale (over 1000 ha) with accurate resource description. In contrast, the remaining 30% of plantation forests are small-scale (less than 1000 ha). It is forecasted that these small-scale forests will supply nearly 40% of the national wood production in the next decade. However, in-depth description of these forests, especially those under 100 ha, is very limited. This research evaluates the use of remote sensing datasets to map and estimate the net stocked plantation area for small-scale forests. We compared a factorial combination of two classification approaches (Nearest Neighbour (NN), Classification and Regression Tree (CART)) and two remote sensing datasets (RapidEye, RapidEye plus LiDAR) for their ability to accurately classify planted forest area. CART with a combination of RapidEye and LiDAR metrics outperformed the other three combinations producing the highest accuracy for mapping forest plantations (user's accuracy = 90% and producer's accuracy = 88%). This method was further examined by comparing the mapped plantations with manually digitised plantations based on aerial photography. The mapping approach overestimated the plantation area by 3%. It was also found that forest patches exceeding 10 ha achieved higher conformance with the digitised areas. Overall, the mapping approach in this research provided a proof of concept for deriving forest area and mapping boundaries using remote sensing data, and is especially relevant for small-scale forests where limited information is currently available.

show abstract

Comparing object-based and pixel-based classifications for mapping savannas

Cited by 269 publications

References 17 publications

Interpretation of Forest Resources at the Individual Tree Level at Purple Mountain, Nanjing City, China, Using WorldView-2 Imagery by Combining GPS, RS and GIS Technologies

Interpretation of Forest Resources at the Individual Tree Level at Purple Mountain, Nanjing City, China, Using WorldView-2 Imagery by Combining GPS, RS and GIS Technologies

An Object-Based Image Analysis Method for Monitoring Land Conversion by Artificial Sprawl Use of RapidEye and IRS Data

Mapping Net Stocked Plantation Area for Small-Scale Forests in New Zealand Using Integrated RapidEye and LiDAR Sensors

Contact Info

Product

Resources

About