A segmentation approach to classification of remote sensing imagery

Kartikeyan, B.; Sarkar, Arnab; Majumder, K. L.

doi:10.1080/014311698215199

Cited by 59 publications

(31 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Given the contribution that spatial attributes can make to land cover classi cation, their increased use is most desirable. Recent interest in an e ective use of spatial and spectral information (Shimabukuro et al 1997, Kartikeyan et al 1998 ) is therefore encouraging. An important consideration in land cover classi cation is consistency and reproducibility.…”

Section: Classi Cationmentioning

confidence: 99%

“…Although numerous algorithms have been developed to quantify spatial relations within images such as texture (Gong et al 1992 ), segment homogeneity (Kartikeyan et al 1998 and references therein) and various others, the spatial Cihlar et al 1998e). In this unsupervised classi cation, steps 1-8 can be carried out in an automated mode but steps 9-10 require analyst's input.…”

Section: Classi Cationmentioning

confidence: 99%

See 1 more Smart Citation

Land cover mapping of large areas from satellites: Status and research priorities

Cihlar¹

2000

International Journal of Remote Sensing

522

220

View full text Add to dashboard Cite

Abstract. Although land cover mapping is one of the earliest applications of remote sensing technology, routine mapping over large areas has only relatively recently come under consideration. This change has resulted from new information requirements as well as from new developments in remote sensing science and technology. In the near future, new data types will become available that will enable marked progress to be made in land cover mapping over large areas at a range of spatial resolutions. This paper is concerned with mapping strategies based on 'coarse' and ' ne' resolution satellite data as well as their combinations. The status of land cover mapping is discussed in relation to requirements, data sources and analysis methodologies-including pixel or scene compositing, radiometric corrections, classi cation and accuracy assessment. The overview sets the stage for identifying research priorities in data pre-processingand classi cation in relation to forthcoming improvements in data sources as well as new requirements for land cover information. Introduction and objectiveLand cover, i.e. the composition and characteristics of land surface elements, is key environmental information. It is important for many scienti c, resource management and policy purposes and for a range of human activities. It is an important determinant of land use and thus of value of land to the society. Land cover varies at a range of spatial scales from local to global, and at temporal frequencies of days to millennia. As the need for environmental planning and management became important, an accompanying call for land cover information emerged in parallel.Land cover mapping is a product of the development of remote sensing, initially through aerial photography (Colwell 1960 ). This is because 'viewing' large areas repeatedly is necessary for acquiring information about land cover. For the same reason, land cover mapping has been perhaps the most widely studied problem employing satellite data, beginning with Landsat 1. However, most of the studies using ' ne' resolution data (i.e. 20-100 m) were methodological in nature, exploring various information extraction techniques and applying these over limited areas. Applications over large areas were hampered by the lack of suitable technology, an absence of a user community with a strong need for such information, a lack of appropriate analysis methodologies, and the cost of data. Thus, at the global level, land cover data sets compiled from ground surveys or various national sources (Mathews 1983, Olson et al. 1983 were, for a number of years, the major source of information.

show abstract

Section: Classi Cationmentioning

confidence: 99%

Section: Classi Cationmentioning

confidence: 99%

Land cover mapping of large areas from satellites: Status and research priorities

Cihlar¹

2000

International Journal of Remote Sensing

522

220

View full text Add to dashboard Cite

show abstract

“…Consequently, the spectral generalisation inherent in pixel-mergers counteracts the salt-and-pepper effect by producing more homogeneous results. This is why segmentation techniques are increasingly used to deal with intensified in-class variability (Kartikeyan et al, 1998;Hill, 1999;Rodrí-guez-Yi et al, 2000;Blaschke et al, 2002;Neubert and Meinel, 2002;Schiewe and Tufte, 2002;Koch et al, 2003;Schneevoigt and Schrott, 2006).…”

Section: Introductionmentioning

confidence: 97%

“…Conventional classification per pixel forms clusters based on spectral similarities alone and can result in many dispersed classes (salt-andpepper effect) which often do not grasp the essence of the information inherent in the scene (Kartikeyan et al, 1998). In contrast, image segmentation prior to classification helps to manage increased geometric resolution by merging adjacent pixels based on grey value homogeneity and form parameters.…”

Section: Introductionmentioning

confidence: 98%

Detecting Alpine landforms from remotely sensed imagery. A pilot study in the Bavarian Alps

et al. 2008

View full text Add to dashboard Cite

“…Image segmentation has been used in land cover classification for several decades [14,[37][38][39][40][41]. The methods available for image segmentation and the applications for OBIA have become more broadly sophisticated over the last decade [12,[42][43][44][45].…”

Section: Image Object Area Training Datamentioning

confidence: 99%

The Effects of Point or Polygon Based Training Data on RandomForest Classification Accuracy of Wetlands

et al. 2015

View full text Add to dashboard Cite

Wetlands are dynamic in space and time, providing varying ecosystem services. Field reference data for both training and assessment of wetland inventories in the State of Minnesota are typically collected as GPS points over wide geographical areas and at infrequent intervals. This status-quo makes it difficult to keep updated maps of wetlands with adequate accuracy, efficiency, and consistency to monitor change. Furthermore, point reference data may not be representative of the prevailing land cover type for an area, due to point location or heterogeneity within the ecosystem of interest. In this research, we present techniques for training a land cover classification for two study sites in different ecoregions by implementing the RandomForest classifier in three ways: (1) field and photo interpreted points; (2) fixed window surrounding the points; and (3) image objects that intersect the points. Additional assessments are made to identify the key input variables. We conclude that the image object area training method is the most accurate and the most important variables include: compound topographic index, summer season green and blue bands, and grid statistics from LiDAR point cloud data, especially those that relate to the height of the return.

show abstract

A segmentation approach to classification of remote sensing imagery

Cited by 59 publications

References 0 publications

Land cover mapping of large areas from satellites: Status and research priorities

Land cover mapping of large areas from satellites: Status and research priorities

Detecting Alpine landforms from remotely sensed imagery. A pilot study in the Bavarian Alps

The Effects of Point or Polygon Based Training Data on RandomForest Classification Accuracy of Wetlands

Contact Info

Product

Resources

About