SOURCE:

Faculty: Environmental Sciences
Department: Surveying & Geo-informatics

CONTRIBUTORS:

Ezeomedo, I. C.
Igbokwe, J. I.

ABSTRACT:

One of the prime prerequisites for better use of land is information on the existing land use patterns and changes in land cover. However, this basic geospatial information is lacking in the study areaand sequel to this, the research aimed at Identification and Mapping of Features in Nnewi, Nigeria, using High Resolution Satellite Image and Support Vector Machine with a view to developing a reliable urban land use/land cover map of the area, which will serve as a base map for land-use planning and monitoring by end-users. The objectives include to: identify and extract Region of Interests (ROIs) in a subset HRSI of the study area, evaluate the result of region of interests and ground truth extraction using the Jeffries-Matusita and Transformed Divergence separability index, perform image classification using Support Vector Machine in ENVI 5.1 Software, evaluate the performance of Support Vector Machine (SVM) and Maximum Likelihood (ML) in mapping geometric features using Error Matrix, Kappa and Post-Confusion Matrix (PoCoMa) template andto produce the landuse and landcover map of Nnewi-North Local Government Area.The methodology includes Image acquisition, enhancement, Sub-setting, ROIs extraction and separability index analysis, supervised classification using SVM and ML, post-processing accuracy assessment, statistical analyses, and preparation of maps. Environment for Visualizing Image (ENVI 5.1) incorporated with Interactive Data Language (IDL 8.3) software, was used for image processing, masking, spatial data analysis and classification. ESRI,ArcGIS 10.2 was employed for database development and production of thematic maps. Microsoft Excel, GraphPad Prism ver.7.0 and IBM SPSS ver.21 were used for statistical analysis. The result of image classification indicates that Nnewi-North local government area is having 13.52% of built-up areas, 24.23% of vegetation, 22.05% of water bodies, Farm lands is 39.40% and open/bare surface is 0.81% using SVM while ML result shows that Built-up Areas is 14.99%, vegetation is 13.01%, water bodies is 34.08%, farm lands is 36.00% and open/bare surface is 1.32%. SVM overall Accuracy is 98.07% and Kappa Coefficient is 0.97 while ML overall accuracy is 82.50% and Kappa coefficient is 0.76. The significance of the mean difference using the “Post Confusion Matrix” (PoCoMa) template showed that the t-statistics is 0.670 with probability value of -0.476 which is greater than 0.05, thus, the null hypothesis was accepted with a deduction that using any of the algorithms (SVM and ML) yields no significance difference in performance and efficiency of result of the map produced. The research revealed that ‘Support Vector Machine Classifier’ is robust in extracting urban landscape from HRSI.Consequently, there is need for periodic urban LULC analysis to guide stakeholders in planning, monitoring and management of urban areas among others.