Hello, I am new to IDL and I would like to create a script that will eliminate noisy pixels from an image and replace them with more valid value. The script will have to do several operations: 1- Verify if each pixel is valid (between +- 2 standard deviation from the mean of the image) 2- For each invalid pixel, calculate the mean of a 9x9 (or other dimensions) convolution mask and replace the value of the pixel by the mean of the mask. So the new value will be calculated from the neighbors. 3- It will be better if the script could calculate the mean of the mask with the valid pixels only inside the convolution mask. I hope the description is clear enought. I already created such a script with Python but the performance was not satisfying... it was really really slow, so I had the idea to try with IDL wich has the repetutation to be performant with matrix. Thank you Max