I’m now detecting change in urban area using TSX data. I preprocessed 3 images in following step;
- Terrain Flattening
- Terrain Correction
However, I found that image colour changed after I coregistered 3 images in one stack. For example:
I didn’t know why this happened, and whether I have done something wrong? Can anyone explain to me?
Thank you in advance!
Which interpolation method did you use? Only Nearest Neighbour conserves original pixel values,
Thank you for your reply!
I use the default parameter setting, and the “Resampling Type” is “NONE”
as stated in the Color Manipulation tab, the statistics are only estimated by a number of pixels. To get the most accurate numbers, you can use the Statistics tool on the raster before and after coregistration.
Another reasons for the changed values might be that the extent of the image changes after resampling it to the extent of the reference image.
It is true that the colour manipulation gives only a rough estimate of the min/max values also the interpolation can cahnge the values.
But both things do not explain a change
Min: 0 -> -338
max 787 -> 1276
Or I don’t understand an essential part of the coregistration.
This is of course possible.
Thank you for your explaination! I redo the coregistration using nearest neighbour resampling method as @mengdahl said, but the colour changing still existed. However, I linear the 2 images(before and after coregistration) to dB, and their Statistics result were very similar, so I think maybe the coregistration result is reliable?
The dB image looked similar and their statistics result were very similar
The color shift still exist:
Another question, for urban change detection application, should I use dB image or the original Gamma0 image?
Thank you in advance!
the color is just a matter of stretch (more obvious in the linearly-scaled data, less visible in the log-scaled data). Please make sure to apply the same minimum and maximum values in the Color Manipulation tab to make the images visually comparable.
If conversion to dB makes sense or not depends on the type of information you want to identify. If you want to separate urban from non-urban land, the linear scaling gives you better contrasts. If you want to investigate changes within the built-up areas, the dB-scaled data might reveal more details.
Thank you for such patient explaination.
I want to investigate changes within the built-up areas, so maybe dB is a better choice.
By the way, is histogram matching a solution to solve the stretch promblem?
the differences are only visual ones, the pixel values are the same.
If you apply the same color stretch to both images, they will look alike
Got it! Thank you for your help!