Next Article in Journal
Monitoring Active Volcanos Using Aerial Images and the Orthoview Tool
Next Article in Special Issue
A Versatile, Production-Oriented Approach to High-Resolution Tree-Canopy Mapping in Urban and Suburban Landscapes Using GEOBIA and Data Fusion
Previous Article in Journal
Delineation of Rain Areas with TRMM Microwave Observations Based on PNN
Previous Article in Special Issue
Object-Based Land-Cover Mapping with High Resolution Aerial Photography at a County Scale in Midwestern USA
Open AccessArticle

Application of In-Segment Multiple Sampling in Object-Based Classification

1
Slovenian Centre of Excellence for Space Sciences and Technologies SPACE-SI, Aškerčeva cesta 12, 1000 Ljubljana, Slovenia
2
Research Centre of the Slovenian Academy of Sciences and Arts, Novi trg 2, 1000 Ljubljana, Slovenia
*
Author to whom correspondence should be addressed.
Remote Sens. 2014, 6(12), 12138-12165; https://doi.org/10.3390/rs61212138
Received: 14 July 2014 / Revised: 11 November 2014 / Accepted: 27 November 2014 / Published: 5 December 2014
(This article belongs to the Special Issue Advances in Geographic Object-Based Image Analysis (GEOBIA))
When object-based analysis is applied to very high-resolution imagery, pixels within the segments reveal large spectral inhomogeneity; their distribution can be considered complex rather than normal. When normality is violated, the classification methods that rely on the assumption of normally distributed data are not as successful or accurate. It is hard to detect normality violations in small samples. The segmentation process produces segments that vary highly in size; samples can be very big or very small. This paper investigates whether the complexity within the segment can be addressed using multiple random sampling of segment pixels and multiple calculations of similarity measures. In order to analyze the effect sampling has on classification results, statistics and probability value equations of non-parametric two-sample Kolmogorov-Smirnov test and parametric Student’s t-test are selected as similarity measures in the classification process. The performance of both classifiers was assessed on a WorldView-2 image for four land cover classes (roads, buildings, grass and trees) and compared to two commonly used object-based classifiers—k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM). Both proposed classifiers showed a slight improvement in the overall classification accuracies and produced more accurate classification maps when compared to the ground truth image. View Full-Text
Keywords: GEOBIA; in-segment analysis; sampling; urban land cover classification; Kolmogorov-Smirnov test statistics; Student’s t-test statistics; probability value GEOBIA; in-segment analysis; sampling; urban land cover classification; Kolmogorov-Smirnov test statistics; Student’s t-test statistics; probability value
Show Figures

Graphical abstract

MDPI and ACS Style

Đurić, N.; Pehani, P.; Oštir, K. Application of In-Segment Multiple Sampling in Object-Based Classification. Remote Sens. 2014, 6, 12138-12165. https://doi.org/10.3390/rs61212138

AMA Style

Đurić N, Pehani P, Oštir K. Application of In-Segment Multiple Sampling in Object-Based Classification. Remote Sensing. 2014; 6(12):12138-12165. https://doi.org/10.3390/rs61212138

Chicago/Turabian Style

Đurić, Nataša; Pehani, Peter; Oštir, Krištof. 2014. "Application of In-Segment Multiple Sampling in Object-Based Classification" Remote Sens. 6, no. 12: 12138-12165. https://doi.org/10.3390/rs61212138

Find Other Styles

Article Access Map by Country/Region

1
Only visits after 24 November 2015 are recorded.
Search more from Scilit
 
Search
Back to TopTop