High Resolution Lunar Image Processing

Page 3

Image enhancement

Our image alignment processes have gone a long way towards removing the negative effects of seeing but at the cost of a soft image. Now the job is to bring out the fine detail that's actually there. The first step in this process is the application of a general  wavelet sharpening in Registax. Before we do that however I need to make an important point. The time to apply wavelets is right after MAP alignment and stacking. If you must do it later then make sure to save the image as TIF or FITS. Do not save as a BMP and then try to apply wavelets at a later time. If you do the generated noise levels will be a lot higher.

My standard wavelet application is made using the linear gaussian filter mode with the wavelet layer sat as shown in the following graphic:

The 1:1 layer is for fine detail enhancement and the rest with the exception of 6:1 are all set to negative values are for noise reduction. Here's the result:

That a substantial improvement but there's still more to do. My next move is to bring the image into PixInsight and resample the image by a factor of 2X using the b-spline function. This can also be done in Iris. Once resampled I bring the image into Photoshop and apply the Focus Magic plug in to bring up that last bit of detail that's lurking in the image. Once that's done the image is resampled back down to it's original size, here's the result:

The last step in the process is to correct the tonal range of the image. We need to open up the shadows without blowing the highlights. You could do this with curves or levels but a far more effective method is the use of contrast masking. Take the link for details of the process. Here's the image after contrast masking has been applied:

Next page, Recap