0

Following is the status of an image that I have:

ExifTool Version Number         : 12.42
File Type                       : TIFF
MIME Type                       : image/tiff
Exif Byte Order                 : Little-endian (Intel, II)
Subfile Type                    : Full-resolution image
Bits Per Sample                 : 8 8 8 8
Compression                     : LZW
Photometric Interpretation      : RGB
Samples Per Pixel               : 4
Rows Per Strip                  : 1
Planar Configuration            : Chunky
Predictor                       : Horizontal differencing

The extra samples count is 0. What does this mean? What is the extra 8 bits present in the data?

Komal
  • 1
  • 1
  • I don't see an extra samples count in your post? – Mark Setchell Jul 04 '22 at 06:47
  • There is no output in exif tool for extra samples count. And I checked the file in notepad++ there was no tag for 338 (152.H) or ExtraSamples. – Komal Jul 04 '22 at 06:50
  • I would be inclined to use **ImageMagick** in the Terminal to separate the 4 channels and lay them out horizontally side-by-side so you can see the channels separately, specifically left-to-right R, G, B and mystery channel `magick YOURIMAGE.TIF -separate +append result.jpg` – Mark Setchell Jul 04 '22 at 06:58
  • The first three images (or channels) are coming as a black and white text image (the original file also looks like a black and white text image and looks identical to the first three) and the fourth image (or channel) is coming as a complete white image. – Komal Jul 04 '22 at 07:23
  • It's probably a pointless transparency channel. If you run `magick identify -verbose YOURIMAGE.TIF` you can look at the statistics for the 4th channel (the one after Blue) and you'll likely see minValue=255, maxValue=255 meaning the entire channel is 255 and holds no useful info. – Mark Setchell Jul 04 '22 at 07:33
  • Thank you. It came out as you said and rightfully so. But I did not understand one thing in the output. Channel depth: Red: 8-bit Green: 8-bit Blue: 8-bit Alpha: 1-bit Shouldn't the Alpha channel be of depth 8-bit here? Since we get that it is a 32 bit image? – Komal Jul 04 '22 at 07:54
  • It is a bit misleading, I agree, but it just means the information in that channel *could* be contained in a single (set) bit although 8 bits are present. I prefer the way `exiftool` expresses it. – Mark Setchell Jul 04 '22 at 07:56

0 Answers0