Sampling scheme
Sample design (where to sample) requires careful thought in terms of producing a representative sample that is practically attainable
- random
- systematic
- stratified
- stratified-random
It is important to ensure that the sampling scheme doesn't coincide with any patterns present on the mapped area. If this is the case, the representativeness of the sample will be compromised.
References:
The most frequently cited references for accuracy assessment are:
Congalton, R. (1991) A review of assessing the accuracy of classifications of remotely sensed data, Remote Sensing of Environment, 37: 35-46.
Stehman, S.V. (2001). Statistical Rigor and Practical Utility in Thematic Map Accuracy Assessment. Photogrammetric Engineering and Remote Sensing, 67(6): p727-734.
Foody, G. (2002) Status of land cover-classification accuracy assessment. Remote Sensing of Environment, 80:185-201.