Print

Print


There's Tika <http://tika.apache.org/>, which has command-line capabilities.  I just launched the UI app, dropped a TIFF on it, and got this output:

Bits Per Sample: 8 8 8 8 bits/component/pixel
Compression: LZW
Content-Length: 262844
Content-Type: image/tiff
Orientation: Top, left side (Horizontal / normal)
Photometric Interpretation: RGB
Planar Configuration: Chunky (contiguous for each subsampling pixel)
Predictor: 2
Rows Per Strip: 30 rows/strip
Samples Per Pixel: 4 samples/pixel
Strip Byte Counts: 20668 7759 13240 15631 14302 17278 11236 14414 6226 5401 7310 4813 12716 5368 4213 3357 5664 6081 8466 12266 8083 8541 14306 7245 11916 9443 4636 705 705 417 bytes
Strip Offsets: 8 20676 28435 41675 57306 71608 88886 100122 114536 120762 126163 133473 138286 151002 156370 160583 163940 169604 175685 184151 196417 204500 213041 227347 234592 246508 255951 260587 261292 261997
Thumbnail Image Height: 881 pixels
Thumbnail Image Width: 1081 pixels
Unknown tag (0x0152): 1
Unknown tag (0x0153): 1 1 1 1
resourceName: tika-view.tiff
tiff:BitsPerSample: 8
tiff:ImageLength: 881
tiff:ImageWidth: 1081
tiff:Orientation: 1
tiff:SamplesPerPixel: 4

	Erik

On Nov 19, 2012, at 14:31 , Kyle Banerjee wrote:

> Howdy all,
> 
> I need to extract all the metadata from a few thousand images on a network
> drive and put it into spreadsheet. Since the files are huge (each is
> 100MB+) and my connection isn't that fast, I strongly prefer to not move
> them before working on them -- i.e. I'm using cygwin and/or windows.
> 
> Just eyeballing these things, I see the headers contain everything I need
> in purty rdf. What's the best way to extract this? I thought tiffinfo would
> do the trick, but it's just giving me technical info. Of course I can just
> parse the files with perl but I'm thinking there just has to be a slicker
> way to do this. What's my best option? Thanks,
> 
> kyle