Conclusion: Classification Methodology

1. Information requirements
   1. Study Area & Data and/or Knowledge: different GIS and Google Earth information and maps
   2. Application problem and required object classes (understanding objects in the study are, size/area distribution of different classes, spectral overlap, etc.)
2. Data acquisition for mapping land cover
   1. New vs. Existing RS data!!
3. Field work
4. Analysis & Basic Assumptions
   1. Selection of image classification method
   2. Available prior knowledge
5. Digital Image Classification
   1. Training stage
   2. Classification
   3. Classification Enhancement (post classification, re-classification, Filters)
6. Classification accuracy assessment
   1. Comparing Classification results and ‘Ground Truth’
   2. calculating various measures of error (e.g., Confusion Matrix)
   3. Challenges in image classification (e.g., mixed pixels).

Ultimately, the material prepares students to understand, process, and interpret remotely sensed data for various applications, such as precision agriculture and environmental monitoring.