Question

1. How to create satellite images able to change detection?

2. How to verify the validity of the data obtained from the satellites?

Answer 1

1) Satellite images are usually used to help us to understand urban expansion or get a broad look at any area of development. Usually satellite images are at medium or large scale. Change detection involves observing the changes in spatial representation of any point by noting it at different times. Change detection methods can be broadly classfied into the following group

a) algebra

b) classification

c) transformation

d) other approaches

The different methods to create satellite images able to change detection are

1) Comparative analysis which creates a difference image (DI)

2) Use machine learing applications with the input of time series data to help in developing automated change detection systems

3) We can combine a multilayer perceptron neural network along with graphically based image segmentation - this will help in very accurate change detection

----------------------------------------------------------------------------------------------------------------------------------------------

Data validity involves checking inpute data to make sure it is consistent with the data requirements of the sytem so that no data errors arise.

In order to verify the validity of our data, we have to check various assumptions at each and every step of the validation process.

In general we can follow the approach below

1) Do a quality check/homogenization of the data

2) Check to see if spatio temporal collocation has been done correctly

3) Check to see if databases have been matched correctly

4) Note down any uncertainties

5) Calculate and tabulate results to determine if your validation meets the minimum guidelines