Application of object-oriented approach in SPOT5 image classification: a case study of interpretation of mudstone in Taiwan

Meng Yu, Zhongming Zhao, Peng ShuProceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2007 · DOI

Abstract

SPOT5 imagery is widely used in urban planning, investigation of land utilization, environmental management etc. for its relatively high-resolution. Object-oriented classification techniques based on image segmentation are being actively studied in the high-resolution image process and interpretation to extract a variety of thematic information. Different from the pixel-based image analysis, the processing of the object-oriented method is based on image segment, not single pixel. The object-oriented classification includes two consecutive processes. An image is subdivided into separated regions according to the spectral and spatial heterogeneity in the image segmentation process. Then the objects are assigned to a specific class according to the class's detailed description in the image classification process. As a case study, the study area is a part of the Taiwan, whose mudstone bare-land is a significant problem due to the poor condition of the soil-physical and microclimate. The SPOT-5 image in March of 2006 is segmented and then these segments are classified to hierarchically linked objects by the eCognition software. By using manual interpreted aerial photo of the same area and traditional classification results as reference for accuracy assessment, this study has higher accuracy compared with the traditional classification.
BibTeX
@inproceedings{yu2007application,
  author = {Meng Yu and Zhongming Zhao and Peng Shu},
  title = {{Application of object-oriented approach in SPOT5 image classification: a case study of interpretation of mudstone in Taiwan}},
  booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
  year = {2007},
  doi = {10.1117/12.747323},
}

Loading related content...

Figure