brad@looking.UUCP (Brad Templeton) (03/25/89)
I would be interested in experiences people have had in scanning pages of text or line drawings and reducing them. Do people have any figures for average sizes of the reduced bitmaps? Tools like FAX use run-length encoding and some huffman to reduce their bitmaps. Are there any tools around that go significantly better. (Other than OCR, of course.) How small does a typical 8.5 by 11 page of stuff get when scanned at 300 by 300 and compressed? I would guess 40K, but I would like to find something that can take it down to 10K by being really smart about what's being scanned. -- Brad Templeton, Looking Glass Software Ltd. -- Waterloo, Ontario 519/884-7473
rick@pcrat.UUCP (Rick Richardson) (03/25/89)
In article <3003@looking.UUCP> brad@looking.UUCP (Brad Templeton) writes: >I would be interested in experiences people have had in scanning pages >of text or line drawings and reducing them. Do people have any figures > >How small does a typical 8.5 by 11 page of stuff get when scanned at >300 by 300 and compressed? I would guess 40K, but I would like to Define 'typical'. An image I have here that is just few line drawings has these stats: TIFF TIFF TIFF SIZE NO CODING PACKBITS LZW figure 2271 x 2540 721,702 60,980 24,290 x 1 bit That seems to be close to your guesstimate when LZW is used inside the TIFF file. But there really isn't much info in this drawing. -- Rick Richardson | JetRoff "di"-troff to LaserJet Postprocessor|uunet!pcrat!dry2 PC Research,Inc.| Mail: uunet!pcrat!jetroff; For anon uucp do:|for Dhrystone 2 uunet!pcrat!rick| uucp jetroff!~jetuucp/file_list ~nuucp/. |submission forms. jetroff Wk2200-0300,Sa,Su ACU {2400,PEP} 12013898963 "" \d\r\d ogin: jetuucp
elt@entire.UUCP (Edward L. Taychert) (03/28/89)
In article <3003@looking.UUCP>, brad@looking.UUCP (Brad Templeton) writes: > How small does a typical 8.5 by 11 page of stuff get when scanned at > 300 by 300 and compressed? I would guess 40K, but I would like to > find something that can take it down to 10K by being really smart about > what's being scanned. It all depends on the source, but fax group 4 format should get you down in your target numbers. For Brad: Copyright 1989, Edward L. Taychert - All rights reserved. Ed Taychert ...!rochester!rocksanne!entire!elt
kevin@jtsv16.UUCP (kevin) (03/29/89)
In article <3349@entire.UUCP> elt@entire.UUCP (Edward L. Taychert) writes: >In article <3003@looking.UUCP>, brad@looking.UUCP (Brad Templeton) writes: > >> How small does a typical 8.5 by 11 page of stuff get when scanned at >> 300 by 300 and compressed? I would guess 40K, but I would like to >> find something that can take it down to 10K > >It all depends on the source, but fax group 4 format should get you >down in your target numbers. > Lets see, 8.5 * 300 = 2550 pixels per scan line, 11 * 300 = 3300 scanlines. Thats 1,051,875 bytes for the image. Taking this down to 10k bytes is about 105:1 compression. Group 4 is not going to do it for you. Ok, maybe if your page is entirely blank. I have done some playing with Group 3 and 4 compression and typical business documents get somewhere between 10:1 and 20:1 compression, depending on the content. You will have to try something else to get the kind of compression you want. I read some time ago that fractal techniques can be used to yield amazing compression factors, but I don't know anything else about it. -- Kevin Brighton kevin@jtsv16.jts.com JTS Computer Systems Ltd. { suncan | geac | uunet }!jtsv16!kevin Toronto, Ontario, CANADA +1 416 665 8910