Flat field correction tool - 10% of my conversions contain strange blotches

[DanielJS97]

I've been using a digital camera with Lightroom and Negative Lab Pro to digitise my collection of 35mm negatives.

I discovered about a year ago that flat field correction is an essential part of film digitization, since no lens/light combination is ever going to be 100% even.

The issue is that flat field correction, after many years, remains buggy. About 10% of my conversions contain strange blotches that aren't visible on either the negative or blank plate. A further 10% become very overexposed and washed out looking. I'm finding that these problems occur within batches rather than individual photos - if one image has an issue, it is likely that all other images corrected with it will contain the same problem.

This is even more annoying since I'm setting FFC to automatically delete the original RAW files upon correction, since I've backed them up separately and wish to save space on my PC. So if a correction fails, I have to manually retrieve the files from a backup drive to reattempt it (note that I have 450 rolls of digitized film, or 16,000 photos, so this is my only option).

The only way that I can get more reliable results is to perform corrections in smaller batches (i.e. 5-10 photos at a time instead of 30-40). This makes no sense though, since LR should be correcting every photo individually in relation to the blank plate, the correction of one photo should have no impact on that of another.

It's also worth noting that I've been experiencing this problem on both Mac and PC over the course of the last 2 years.

This is quite annoying since FFC correction is a powerful tool with spotty implementation - I'm hoping Adobe can find a fix to make it more reliable in the near future.

[mcknig_k]

I replied below on another bug report.  What is happening is Lightroom is using non-cabliration frames and applying them to your photos, rather than using the cablibration frame.  The plugin seems to decided what is a calibration frame and what isn't, rather than the user telling the program what's what.  Here are my findings with the issue, as replied on another bug report:

"I am experiencing a similar problem, although I didn't put 2+2 together that it occurred after the update.  Previously, I was able to easily apply FFC to a large number of negatives with a single calibration frame.  Now, if I try to do a large group, Lightroom will grab other non calibration frames and use them for calibration, creating ghost images in the .dng files.  Another issue seems to be that trying to do a FFC on negative that even slightly resemebles a calibration frame (e.g., an open landscape), Lightroom seemingly does not know which one is the calibration frame, so does nothing.  This is a major issue as FFC is essentially unusable right now and is important for the workflow of many people doing film scanning.  The FFC needs to be able to do the following:

1) the user should be able to tell Lightroom which frame is the calibration frame rather than the program making a guess and frequently getting it wrong.

2) the user should be able to force a FFC for any frame, rather than Lightroom mistaking a photo needing correction for another calibraiton frame and doing nothing

3) the user should be able to apply FFC to a large number of frames without it resulting in ghost images as Lightroom uses the wrong frame for calibration"

I am new to reporting here - does Adobe review and respond to user requests here, or are we screaming into the void?