In the process of image classification which of the following methods result in a greater accuracy of classes within an image actually matching land use patterns on the ground ?
MH SET - 2023, SET Exam - 2023,
Answer: C
The more accurate method is Supervised classification.
In supervised classification the analyst first chooses training areas whose land use is known from field checks or maps, and the computer assigns every pixel to the class it most resembles. Because the classes are defined from ground truth, the result matches land use on the ground more closely.
Unsupervised classification groups pixels into spectral clusters automatically, and the analyst only labels them afterwards, so its classes need not correspond to land-use categories. 'Automated' and 'Robotic' are not classification methods in this sense.
Please log in or register to report this question.