| |
What is Change Detection?
"Change Detection" is the process through which we can find the differences between two images.
What is it used for?
As nowadays, satellite imaging has become so popular, the specialists are continuously working on these images to improve and control the environmental parameters, like finding the changes in woods & plant areas being developed or ruined, rivers & lakes coast, earth excavations as a result of city enlargements, weather researches and so on.
Classic Algorithms:
In classic algorithm the first images is subtracted from the second one. This ensures that the un-changed pixels will get zero-color (white/black color), while the changed pixels will get highlighted by a (difference-value) color.
| |
 |
| |
|
|
| Well, the classic method is fairly good and applicable; but, let's see how we can add some spice to it! |
My Method:
In my method the program does a Morph between the first and the second images. Morphing between images makes an animated gradual mid-steps images by which the user can visually see the changes.
Advantages:
According to the algorithm discussed above, the advantages can be as the followings:
- This method is more natural for the brain to perceive.
- The result of such a change detection is not a still image, but an animated one, which interactively and successively shows the differences.
- It helps the user not lose the overall imagination of the district (in Satellite Images), because the classic method does not keep the whole two images in the result image.
Disadvantages:
- Because of making more than one image, the process takes longer than the classic one.
- It might be not so useful in cases where the changes are trivial, because of the times being close.
- It cannot be used as an input to another system; it is perceived by human only.
Possible Improvements/Developments:
To improve this algorithm, by mixing the classic method with this algorithm, it is can be developed in a way so that it helps the user automatically zoom into the points that are very likely to have changes (even trivial ones).
|
|