Mapping East African tropical forests and woodlands — A comparison of classifiers

Nangendo, Grace; Skidmore, Andrew K.; Oosten, Henk van

doi:10.1016/j.isprsjprs.2006.11.003

Cited by 33 publications

(13 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Use of these ancillary spatial data and expert rules resulted in greater discrimination of four cover classes of L. camara which might otherwise have been difficult to classify using remotely sensed data alone. This study corroborates the work of other researchers, who also combined conventional classification methods (e.g., the maximum likelihood and spectral angle mapper) with an expert system to map vegetation cover and reported significant increases in mapping accuracy (Nangendo et al, 2007; Bars represent means and whiskers represent 95% confidence intervals for each cover class (n = 31, 30, 30, and 32 for the absent, low, medium, and high cover classes, respectively). Significant differences in spectral values between L. camara cover classes are denoted by different letters based on Scheffe's post hoc pairwise comparison tests (P < 0.05).…”

Section: Discussionsupporting

confidence: 90%

Integrating conventional classifiers with a GIS expert system to increase the accuracy of invasive species mapping

Masocha

Skidmore

2011

International Journal of Applied Earth Observation and Geoinfor

Self Cite

View full text Add to dashboard Cite

Section: Discussionsupporting

confidence: 90%

Integrating conventional classifiers with a GIS expert system to increase the accuracy of invasive species mapping

Masocha

Skidmore

2011

International Journal of Applied Earth Observation and Geoinfor

Self Cite

View full text Add to dashboard Cite

“…This is valid for both the optical [19]- [25] and L-and C-band SAR data when several image acquisitions are available from the same location [59]- [63]. The land cover and forest cover classification accuracies range from 50% to 80% depending on the number of classes [64]. In a multi-class classification, the performance of SAR data has been poorer than with optical data [65].…”

Section: B Comparison With Other Studiesmentioning

confidence: 78%

Improved Mapping of Tropical Forests With Optical and SAR Imagery, Part I: Forest Cover and Accuracy Assessment Using Multi-Resolution Data

Häme

Kilpi

Ahola

et al. 2013

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

This paper describes an improved concept for the mapping of tropical forest classes with ALOS AVNIR and ALOS PALSAR data. The improvement comes from a combination of a sample of very high resolution (VHR) satellite images with medium resolution wall-to-wall mapping in a statistical sampling framework. The approach developed makes it possible to obtain reliable information on mapping accuracy over the whole area of interest.A simulation study indicated that the sample of VHR images should be collected in a stratified manner using small (25 km ) images. The VHR images should cover approximately one percent of the total area of interest, depending on the accuracy requirement. The recommended size of the reference plots (population units) that are selected within the VHR imagery is in the order of 50 m by 50 m. In a systematic selection the plots should be located at a distance of several hundred meters from each other.The forest variables were predicted with an unsupervised fuzzy classification method. The ALOS AVNIR-based forest/non-forest mapping accuracies varied between 68% and 97% of the areas of the VHR images. The corresponding ALOS PALSAR mapping accuracies were poorer. At AVNIR resolution, the area of natural forest was over-estimated, and the degree of disturbance underestimated in humid, heavily disturbed parts of the study area in Laos. The three predictions for the total forest fraction from VHR, AVNIR and PALSAR data over the area that was covered by the VHR images were 55.1%, 53.6%, and 52.8%, respectively.

show abstract

“…When the spectral angle between the pixel and the reference object is less than the given threshold value, the pixel can be considered in the same class as the reference object. The SAM combined with different vegetation distribution rules can be adopted to improve the classification accuracy of six species of forest-woodland-savannah mosaic, and compared with the maximum likelihood classification (MLC); the classification accuracy was increased by 10% in comparison to 85.2% for MLC [52]. By removing spectrum noise or making differential spectrum, SAM can be adopted to eliminate the influence of the vegetation environmental background and the phenomenon caused by the same object with different spectrums to improve the vegetation identification precision.…”

Section: Land Use/land Cover Classificationmentioning

confidence: 99%

Assessment Impacts of Weather and Land Use/Land Cover (LULC) Change on Urban Vegetation Net Primary Productivity (NPP): A Case Study in Guangzhou, China

Zhao

et al. 2013

Remote Sensing

View full text Add to dashboard Cite

Net primary productivity (NPP) can indicate vegetation ecosystem services ability and reflect variation response to climate change and human activities. This study applied MODIS-1 km NPP products to investigate the NPP variation from 2001 to 2006, a fast urban expansion and adjustment period in Guangzhou, China, and quantify the impacts of weather and land use/land cover (LULC) changes, respectively. The results showed that the NPP mean value increased at a rate of 11.6 g·C·m g·C. The spatiotemporal of NPP varied obviously in the central area, suburb and exurb of Guangzhou driven by three patterns of weather and LULC changes. By the interactive effects and the weather variation dominated effects, NPP of most areas changed slightly with dynamic index less than 5% of NPP mean value in the central area and the suburb. The LULC change dominated effects caused obvious NPP reduction, by more than 15% of the NPP mean value, which occurred in some areas of the suburb and extended to the exurb with the outward urban sprawl. Importantly, conversion from wood grassland, shrublands and even forests to croplands occupied by urban landscapes proved to be a main process in the conversion from high-NPP coverage to low-NPP coverage, thereby leading to the rapid degradation of urban carbon stock capacity in urban fringe areas. It is helpful for government to monitor urban ecological health and safety and make relevant policies.

show abstract

Mapping East African tropical forests and woodlands — A comparison of classifiers

Cited by 33 publications

References 27 publications

Integrating conventional classifiers with a GIS expert system to increase the accuracy of invasive species mapping

Integrating conventional classifiers with a GIS expert system to increase the accuracy of invasive species mapping

Improved Mapping of Tropical Forests With Optical and SAR Imagery, Part I: Forest Cover and Accuracy Assessment Using Multi-Resolution Data

Assessment Impacts of Weather and Land Use/Land Cover (LULC) Change on Urban Vegetation Net Primary Productivity (NPP): A Case Study in Guangzhou, China

Contact Info

Product

Resources

About