Importing TIF Images

Usage & Issues
, , ,
#1

Hi, I’m trying to open a TIFF images that was converted from RAW in Fiji. It was taken with a red lens filter so the image should have a red hue. When I open it in Fiji, it has a blue hue instead. Furthermore, it opens as an image with 3 RGB channels all of the same image. If I open the same image in GIMP, it will display correctly as one image. I’ve tried converting the image to an RGB Image to collapse the stacks, but it keeps its blue hue. Anyone know how I can import the TIFF image correctly?

Image in question: https://drive.google.com/file/d/0B7T-JRzb9SRbU0hjbkF6VFpXRFE/view?usp=sharing
ScreenShots: http://imgur.com/a/rSqTb
Top screenshot is Fiji import (I use File>Open)
Bottom image is Windows Photo Viewer. This is how I want it imported to Fiji.

#2

I can’t access your Google Drive image, but I would recommend checking Image> Adjust > Brightness & Contrast. Set Minimum = 0 and Maximum = 255 for all 3 channels (Red, Green, Blue). Sometimes images are auto-contrasted upon opening.

#3

I apologies, the link should now work. I checked the Min and Max values and they are both set to 0 and 255, respectively for all 3 channels.

#4

Thanks for fixing the link. I was able to get the image displayed correctly by setting Min and Max to the full 16-bit range.

To do this, go to Image > Adjust > Brightness & Contrast and click Set, then set options as shown in the screenshot. Although the image window below says “Red” the screenshot is showing the composite with all channels merged.

864×722 512 KB

If you now do Image > Type > RGB Color you will get a non-composite RGB image that preserves the original color balance.

The reason it looked wrong at first is that when the image is opened, the display is auto-contrasted (Min and Max are changed) based on each channel’s range of pixel values. This causes the dim blue and green channels to be enhanced in the display, even though the actual pixel values are lower than in the red channel.

My previous suggestion didn’t work because the image is 16 bits per channel rather than 8 bits per channel as I had guessed.

Hope this helps.

2 Likes