Best Flatbed Scanner for Scanning Books: A continuation of "DocuMate 252/262 or fi-4120C2 for archiv

  • Thread starter Thread starter renethx
  • Start date Start date
R

renethx

This article is a continuation of my previous article titled
"DocuMate 252/262 or fi-4120C2 for archiving documents?" I posted
on Feb 9, 2005 in this group. I would like to add several tips in
digitizing paper documents and archiving them I noticed in course of
doing so in my personal library.

Compression methods for color images

Saving color images in lossless compression format needs lots of disk
space and is not practical. The old compression method of JPEG (high
quality, say, Quality 90 and Sub-sampling 1:1 in ThumbsPlus 7) is still
very good and reduces the file size 5 to 10 times. Degradation of the
image quality is hardly noticeable unless you magnify and examine it
carefully.

OCR programs

Right now the best OCR program is OmniPage 15 Professional from
ScanSoft. I have been using it for a month and I am very impressed. OCR
results are very accurate. It offers full customization for output PDF
files. It comes with a modest set of image enhancement tools. A list of
enhancement steps can be saved as a template and used in any other
workflow later. Image never deforms in the deskew process. It allows
highly customizable workflows and jobs. It allows multiple instances
and can do two or more tasks simultaneously in the same system (you
will need a dual-core processor to use multiple instances effectively,
however). There are a few bugs. In particular the program is not stable
and often shuts down abruptly with no apparent reason. Eventually they
will be fixed in service packs in future (I hope).

I also tried FineReader 8 Professional, but it was much lower than my
expectation. As far as I tested there is no big improvement over
Version 6. It does not allow batch jobs (well, OP Professional is
roughly equivalent to the much pricier FR Corporate edition), image
deforms in deskew process, poor support for output PDF formats (a step
back from FR6), etc.

Flatbed scanners

I often scan bound books because many old academic books I need are out
of print and the only way of keeping it on hand is borrow it from a
library and scan the entire book with a flatbed scanner. For this
purpose I have been using Canon CanoScan LiDE 80 and have been very
satisfied with its performance. Recently I tested a bunch of new models
of flatbed scanners to determine the best scanner for scanning books.
The most important factor is the scan speed. (10-second difference in
scanning a single page will result in more than 1-hour difference in
scanning a 500-page book.) Image quality is also important if you scan
color documents and pictures too. Ergonomics is another important
factor to scan a book quickly. I tested Canon CanoScan 8400F, 4200F,
LiDE 80, LiDE 60, Epson Perfection 3490 and Visioneer OneTouch 9420.

Speeds

To measure scan speeds, I used OP15's TWAIN interface (with
"Automatically scan pages" ON and "Time between scans" 0 sec),
scanned a letter-size document ten times continuously and took the
average scan time per page. Scan time heavily depends on the CPU speed
and the memory bandwidth. For example, scanning a letter-size document
with 4200F in Color at 600dpi takes

31 sec with Athlon 64 3200+ @2630MHz (overclocked),
35 sec with Athlon 64 3200+ @2000MHz (default clock speed),
48 sec with Athlon XP 2400+ @2163MHz (overclocked; default is 2000MHz)

The processor used in this test is Athlon 64 3200+ @2630MHz. The
following is the results of measurement at 400di and 600dpi in each
mode. Unit is second.

CanoScan 8400F: B&W: 17, 29, Grayscale: 18, 33, Color: 19, 35.
CanoScan 4200F: B&W: 15, 26, Grayscale: 16, 30, Color: 16, 31.
CanoScan LiDE 80: B&W: 22, 22, Grayscale: 22, 22, Color: 46, 46.
CanoScan LiDE 60: B&W: 34, 34, Grayscale: 34, 34, Color: 53, 53.
Perfection 3490: B&W: 29, 38, Grayscale: 29, 42, Color: 29, 47.
OneTouch 9420: B&W: 28, 28, Grayscale: 31, 44, Color: 32, 34.

LiDE 80 is by far the fastest in both B&W and Grayscale at 600dpi. For
color documents, 4200F is the fastest at 600dpi.

Image Quality

In B&W mode, there is no problem with any of the scanners. Color images
are very good with any of the Canon and Epson scanners, though color
correction may be necessary for some of them. For example, images from
LiDE 80 are reddish at all ranges (with either "Recommended" or
"CanoScan LiDE 80 Reflective (W) - sRGB IEC61966-2.1" color
management). Like other Visioneer scanners, OneTouch 9420 sets the
white point around 230 by default and there is no way to change it. As
a consequence, the information of lighter parts of the image is lost
permanently. This scanner is unsuitable for pictures for this reason.
Even for color documents, light colors look washed out with this
scanner.

Ergonomics

The reference point of the plate in both CanoScan 8400F and 4200F is
located at the right back corner and it is very difficult to align a
page to it correctly. These scanners should be avoided to scan a bound
book. CanoScan 8400F is also very bulky (nearly three times larger than
LiDE 80 in volume). Both LiDE 80 and LiDE 60 are very compact, easy to
handle and suitable for scanning a bound book. The position of scan
buttons is not good, however. (Cosmetically LiDE 80 made of aluminum is
much better that LiDE 60 made of plastic.) Epson is also good. As for
OneTouch 9420, you need to remove the lid to scan a book and the scan
buttons cannot be used because they are covered by the other side of
the book pages.

Conclusion

The best way to scan a bound book in B&W mode is to use Canon LiDE 80
with OP15's TWAIN interface (with Mode Grayscale, Size B5 in most
cases, "Automatically scan pages" ON and "Time between scans" 0
sec), then save to TIFF files, and change Color Depth to Bi-level (1
bit) with ThumbsPlus (with Threshold 450 for darker documents, 500 for
normal and 550 for lighter documents; these are roughly equivalent to
Threshold 84, 96 and 108 respectively in CanoScan ScanGear). The reason
for scanning in Grayscale mode first is that the deskew process of OP15
works more elegantly with Grayscale than B&W. In this way I am able to
scan a 500-page book in less than 3 hours at 600 dpi in B&W and the
result is a near perfect copy of the entire book.

BTW Canon LiDE 80 has been discontinued. Fortunately it appears on eBay
frequently. If you search in Froogle, you may find LiDE 80 at several
retailers, but I am pretty sure that none of them actually has it in
stock. These retailers never say availability explicitly online to
attract customers and sometimes try to sell a much cheaper model at the
same price of LiDE 80. You had better stay away from them to save time
and money.
 
I save magazine articles in descreened 150 dpi Photoshop Elements ver
1 JPEG #8 compression using an Epson 2400 Photo. The artifacts are
bearable, as is the file size. I don't care about OCR or search
capabilities in the articles. Chances are high that I may never
revisit the articles but if want to, they will be waiting on the hard
drive or a DVD.

Unlike your book situation, I physically separate the individual
magazine pages, index them into the corner of the flatbed, and scan
with the lid up. Deskew is not an issue. Book page positioning is
difficult at best.

Leaving the scanner lid up seems to show less of the "ghost print"
from the other side of a scanned magazine page, which again would not
be a problem with a typical book page.

Your experience is very interesting. Thanks for the review and
comparison of scanner data.
 
----- Original Message -----
From: "renethx" <[email protected]>
Newsgroups: comp.periphs.scanners
Sent: Sunday, October 16, 2005 22:43
Subject: Best Flatbed Scanner for Scanning Books: A continuation of
"DocuMate 252/262 or fi-4120C2 for archiving documents?"

OCR programs

Right now the best OCR program is OmniPage 15 Professional from

I also tried FineReader 8 Professional, but it was much lower than my
expectation. As far as I tested there is no big improvement over
Version 6. It does not allow batch jobs (well, OP Professional is
roughly equivalent to the much pricier FR Corporate edition), image
deforms in deskew process, poor support for output PDF formats (a step
back from FR6), etc.


I have to ask if you actually used FineReader correctly. I haven't used 15,
but OmniPage14 sucked.

To use FR you have to use the FR scanner interface through either wia or
twain depending on your scanner. NOT the scanner sw interface through the
twain driver. This gets you magic b/w scans. OmniPage14 sucked at
correcting the light/dark stripes across the gutter of a book scan.
The next biggest usability plus is the customizable scan size - you can
easily set it to just the body (exclude the header's and footers). This
saves a lot of work removing page numbers later.

No Batch mode?
You can scan until you run out of paper (multipage scan mode, under options
it's in a wierd place) you can recognize in the background while scanning or
you can recognize all pages at once. Since it does brightness correction at
scan time what else would you be "batch" processing?

The really useful new feature in 8 is the correction for the distortion in
the gutter of a book when doing a doublepage scan. This seams to only work
well in portrait mode though.
 
Robert ã®ãƒ¡ãƒƒã‚»ãƒ¼ã‚¸:
----- Original Message -----
From: "renethx" <[email protected]>
Newsgroups: comp.periphs.scanners
Sent: Sunday, October 16, 2005 22:43
Subject: Best Flatbed Scanner for Scanning Books: A continuation of
"DocuMate 252/262 or fi-4120C2 for archiving documents?"




I have to ask if you actually used FineReader correctly. I haven't used 15,
but OmniPage14 sucked.

I have used FR6 Pro for two years correctly (I believe). I used FR8 Pro
for 30 days and requested a refund because there is no improvement over
FR6, at least with regard to my specific needs. (Please read my article
"DocuMate 252/262 or fi-4120C2 for archiving documents?" to understand
my needs.) I have used OP12 Pro that came with DocuMate 262 for a years
and it sucks. I haven't used OP14 Pro, but from what I read in various
user groups, it also sucks. I tried OP15 Pro (paid only $99 as I bought
it within 30 days of the release) and was extremely impressed. It is a
totally different animal from OP12 Pro, and perhaps OP14 Pro.
To use FR you have to use the FR scanner interface through either wia or
twain depending on your scanner. NOT the scanner sw interface through the
twain driver. This gets you magic b/w scans. OmniPage14 sucked at
correcting the light/dark stripes across the gutter of a book scan.
The next biggest usability plus is the customizable scan size - you can
easily set it to just the body (exclude the header's and footers). This
saves a lot of work removing page numbers later.

If you find FR very good, that's fine and I have no objection. For me,
deformation of images in deskew process in FR6/8 Pro (whether I load
files or scan images using FR interface) is absolutely unacceptable.
Nowadays, even a cheep image management program like ThumbsPlus gives
perfect rotation of images.

I recommend those who want to buy an OCR program to try both FR8 Pro
and OP15 Pro and keep the better one (for you). You can request a
refund within 30 days whatever the reason is. Refund process in Digital
River (through which you will buy both FR and OP) is completely
automated and there will be no hassle. You can buy Upgrade versions of
both even if don't have a previous version (installation process will
never check a previous version.)
No Batch mode?
You can scan until you run out of paper (multipage scan mode, under options
it's in a wierd place) you can recognize in the background while scanningor
you can recognize all pages at once. Since it does brightness correctionat
scan time what else would you be "batch" processing?

The really useful new feature in 8 is the correction for the distortion in
the gutter of a book when doing a doublepage scan. This seams to only work
well in portrait mode though.

A simplest example of batch processing I have in my mind is the
following. Suppose that I have 1000 pdf files (image only; mostly old
jurnal articles) in a folder, say, C:\tmp1 and I want to OCR all the
files in this folder and save to pdf files (image on text) in the
folder C:\tmp2. With OP15 Pro I can create a Workflow (a series of
processes) to load, OCR, and save files automatically. I can even
schedule this process at a specific time. All what I have to do is just
click a mouse and wait until the entire process finishes (it may take a
whole day, though). Adobe Acrobat 6.0/7.0 Professional has a similar
capability. Unfortunately FR6 Pro cannot do such a process. I have to
load each file manually, wait for OCR and have to specify the file name
and location to save. Doing this for 1000 files manually is absolutely
unpractical. Maybe FR6/8 Corporate can do such a process (I don't know
though, as I have never used it), but it is much pricier than OP15 Pro.
 
Robert ??????:
----- Original Message -----
From: "renethx" <[email protected]>
Newsgroups: comp.periphs.scanners
Sent: Sunday, October 16, 2005 22:43
Subject: Best Flatbed Scanner for Scanning Books: A continuation of
"DocuMate 252/262 or fi-4120C2 for archiving documents?"




I have to ask if you actually used FineReader correctly. I haven't used
15,
but OmniPage14 sucked.

- I have used FR6 Pro for two years correctly (I believe). I used FR8 Pro
- for 30 days and requested a refund because there is no improvement over
- FR6, at least with regard to my specific needs. (Please read my article
- "DocuMate 252/262 or fi-4120C2 for archiving documents?" to understand
- my needs.) I have used OP12 Pro that came with DocuMate 262 for a years
- and it sucks. I haven't used OP14 Pro, but from what I read in various
- user groups, it also sucks. I tried OP15 Pro (paid only $99 as I bought
- it within 30 days of the release) and was extremely impressed. It is a
- totally different animal from OP12 Pro, and perhaps OP14 Pro.
To use FR you have to use the FR scanner interface through either wia or
twain depending on your scanner. NOT the scanner sw interface through the
twain driver. This gets you magic b/w scans. OmniPage14 sucked at
correcting the light/dark stripes across the gutter of a book scan.
The next biggest usability plus is the customizable scan size - you can
easily set it to just the body (exclude the header's and footers). This
saves a lot of work removing page numbers later.

- If you find FR very good, that's fine and I have no objection. For me,
- deformation of images in deskew process in FR6/8 Pro (whether I load
- files or scan images using FR interface) is absolutely unacceptable.
- Nowadays, even a cheep image management program like ThumbsPlus gives
- perfect rotation of images.


But the subject of this thread is "scanning books", not converting pdf's.

OP didn't do custom page size scans - only letter sized (2.5x as long for
scanning paperbacks) Plus you have to manually remove the title and page
numbers after OCR conversion - lots of manual work.

OP didn't do the de-paging of 2 page book scans and subsequent alignment of
each page or the magic brightness adjustment of FR. And I do mean magic.
It lets you get usable b/w scans of books. Every other program I've tried
couldn't deal with the varying brightness across the gutter. OP requires
grayscale or color scans of books. Otherwise you get a big black stripe
down the gutter.

Now if the comparison is FR 6 vs 7 vs 8, no I wouldn't recommend anyone pay
for the upgrade, they are all just about equally good at scanning books.
 
Robert ã®ãƒ¡ãƒƒã‚»ãƒ¼ã‚¸:
- If you find FR very good, that's fine and I have no objection. For me,
- deformation of images in deskew process in FR6/8 Pro (whether I load
- files or scan images using FR interface) is absolutely unacceptable.
- Nowadays, even a cheep image management program like ThumbsPlus gives
- perfect rotation of images.


But the subject of this thread is "scanning books", not converting pdf's.

"pdf's"??? Here I am talking about TIFF or BITMAP images (or whatever
image file format FR accepts if I load files) or images directly
scanned from the scanner by FR interface. Where did you find the word
"pdf" in the above quotation? Read carefully.
OP didn't do custom page size scans - only letter sized (2.5x as long for
scanning paperbacks)

WRONG! OP15 allows Letter, A4, A5, B5, B6. A5 (about 6" x 8") or B6 (5"
x 7") is small enough for paperbacks. BTW scanning paperbacks with
flatbed scanner is completely a waste of time (assuming paperbacks are
cheap and easily obtained unlike hard-to-find academic books). Use a
duplex sheetfed scanner instead!
Plus you have to manually remove the title and page
numbers after OCR conversion - lots of manual work.

How do you know that I remove title and page numbers? I never remove
them. My final goal is keep the EXACT copy of the book, including title
and page numbers.
OP didn't do the de-paging of 2 page book scans and subsequent alignment of
each page or the magic brightness adjustment of FR. And I do mean magic.
It lets you get usable b/w scans of books. Every other program I've tried
couldn't deal with the varying brightness across the gutter. OP requires
grayscale or color scans of books. Otherwise you get a big black stripe
down the gutter.

Again I never do 2-page book scan. Most books I scan are around 7" x
10". To obtain PERFECT images in B&W (or even Color) mode, THE only way
is first scan with OP15 interface in Grayscale (or Color) mode (OP15 is
THE only program that deskews document images perfectly), convert to
TIFF or BITMAP images, then edit images by an image editing program
(like PhotoShop) and finally convert to 1-bit images. FineReader 6/7/8
always DEFORMS images UGLILY in deskew process (whether images are
loaded from files or scanned by FR interface or the scanner's interface
in any mode) and resulting images are unusable (at least for me)!!
Now if the comparison is FR 6 vs 7 vs 8, no I wouldn't recommend anyone pay
for the upgrade, they are all just about equally good at scanning books.

Why do you use the word "no"? (Cheap rhetoric!) What I wrote is:
- I recommend those who want to buy an OCR program to try both FR8 Pro
- and OP15 Pro and keep the better one (for you). You can request a
- refund within 30 days whatever the reason is. Refund process in Digital
- River (through which you will buy both FR and OP) is completely
- automated and there will be no hassle. You can buy Upgrade versions of
- both even if don't have a previous version (installation process will
- never check a previous version.)

I NEVER recommended to upgrade from FR6/7 to FR8! Nobody shouldn't pay
for the upgrade to FR8 if he/she has already FR6 or FR7 simply because
FR8 is NOT WORTH the upgrade. Moreover ALL the versions are all just
about EQUALLY BAD at scanning books IN THAT they deform images uglily.
 
Robert ã®ãƒ¡ãƒƒã‚»ãƒ¼ã‚¸:
OP didn't do the de-paging of 2 page book scans and subsequent alignment of
each page or the magic brightness adjustment of FR. And I do mean magic.
It lets you get usable b/w scans of books. Every other program I've tried
couldn't deal with the varying brightness across the gutter. OP requires
grayscale or color scans of books. Otherwise you get a big black stripe
down the gutter.

Although I have never done 2-page scanning, I picked up a small
paperback and tried it with both FR6 and OP15 to see what you are
talking about. The settings are:

Scanner: Canon CanoScan LiDE 80

FR6
Use FineReader Interface: ON, Split dual pages: ON, Detect image
orientation (during recognition): ON, Picture scanning mode:
Black-&-white pictures, Resolution: 600dpi, Brightness: Automatic,
Papersize: Letter

OP15
Tools > Options > Scanner > General > Mode: B&W, Resolution: 600 DPI,
Size: Letter, Source: Flatbed, Brightness: Automatic.
Tools > Options > Process > Image preprocessing > Deskew image ON
Tools > Options > Process > Recognition > Look for facing pages ON

After scanning, I OCRed images and saved to a pdf file (image on text).
The results with both programs are satisfactory. Both split pages and
rotate images correctly. But a black strip appears across the gutter in
either case. The strip is much broader with FR6 than OP15. Apparently
Deskew does not work with OP15 in spite of the fact that Deskew was
checked (an ugly bug of OP15!). Deskew works with FR6, but PDF page
size differs considerably page by page because of different rotation
angle for each page, and letters are deformed (paper quality of
paperbacks is in general very low and deformation of letter shapes may
not a big problem, though). I don't understand what is so "magical"
with FR6.

Well, I will never do 2-page scanning with either FR6 or OP15 even if I
have to scan paperbacks with a flatbed scanner because the resulting
pdf file is just too ugly to read. If I saved to text documents (Word
document, plain text etc.), then the results would be perfect with both
programs, however. That's my personal opinion. Everyone has his/her own
needs, sense of beauty, and opinion.

Your biggest problem is that you are discussing OP's problems without
actually using the current version OP15. I wrote NOT "the best OCR
program is OmniPage 14" BUT "the best OCR program is OmniPage 15" in my
original message. If you want to discuss goodies of FR, that's fine,
but you cannot discuss OP without actually using it (I mean, the
current version). All of my reviews and comparisons of sheedfed/flatbed
scanners are based on my actual use of them, needless to say!!
 
- If you find FR very good, that's fine and I have no objection. For me,
- deformation of images in deskew process in FR6/8 Pro (whether I load
- files or scan images using FR interface) is absolutely unacceptable.
- Nowadays, even a cheep image management program like ThumbsPlus gives
- perfect rotation of images.


But the subject of this thread is "scanning books", not converting pdf's.



-"pdf's"??? Here I am talking about TIFF or BITMAP images (or whatever
-image file format FR accepts if I load files) or images directly
-scanned from the scanner by FR interface. Where did you find the word
-"pdf" in the above quotation? Read carefully.


"A simplest example of batch processing I have in my mind is the
following. Suppose that I have 1000 pdf files (image only; mostly old
jurnal articles) in a folder, say, C:\tmp1 and I want to OCR all the
files in this folder and save to pdf files (image on text) in the
folder C:\tmp2. With OP15 Pro I can create a Workflow (a series of"

OP didn't do custom page size scans - only letter sized (2.5x as long for
scanning paperbacks)

- WRONG! OP15 allows Letter, A4, A5, B5, B6. A5 (about 6" x 8") or B6 (5"
- 7") is small enough for paperbacks. BTW scanning paperbacks with
-flatbed scanner is completely a waste of time (assuming paperbacks are
-cheap and easily obtained unlike hard-to-find academic books). Use a
-duplex sheetfed scanner instead!

FR does custom paper size. Not scanning the crap off the page helps a lot with the OCR.
Most duble page paperbacks are about 7.5x5.9
Plus you have to manually remove the title and page
numbers after OCR conversion - lots of manual work.

- How do you know that I remove title and page numbers? I never remove
- them. My final goal is keep the EXACT copy of the book, including title
- and page numbers.

How do you read them? On any normal reader (uBook, MobiPocket Reader, or even MS Reader) I want the text to flow to fit the reader screen size - not the arbitray shape of the source.

OP didn't do the de-paging of 2 page book scans and subsequent alignment of
each page or the magic brightness adjustment of FR. And I do mean magic.
It lets you get usable b/w scans of books. Every other program I've tried
couldn't deal with the varying brightness across the gutter. OP requires
grayscale or color scans of books. Otherwise you get a big black stripe
down the gutter.

- Again I never do 2-page book scan. Most books I scan are around 7"x10".

Most Letter&A4 scanners will do 8.5x12.5 which is fine for most hardback books.

- To obtain PERFECT images in B&W (or even Color) mode, THE only way
- is first scan with OP15 interface in Grayscale (or Color) mode (OP15 is
- THE only program that deskews document images perfectly), convert to
- TIFF or BITMAP images, then edit images by an image editing program
- (like PhotoShop) and finally convert to 1-bit images.

FR 7 & 8 (I haven't used 6) do usable bw scanns for OCR. If you're trying
to capture pictures with the text then you have to scan in grayscale or color,
but NOT for OCR with FR.


- FineReader 6/7/8
- always DEFORMS images UGLILY in deskew process (whether images are
- loaded from files or scanned by FR interface or the scanner's interface
-in any mode) and resulting images are unusable (at least for me)!!

Unusable(?) They OCR just fine.
If you are trying to capture illistrations and not just ocr text thats different.
Now if the comparison is FR 6 vs 7 vs 8, no I wouldn't recommend anyone pay
for the upgrade, they are all just about equally good at scanning books.

- Why do you use the word "no"? (Cheap rhetoric!)

I was disagreeing with *MY* FR8 recommendation.
If the compaison is OP15 vs FR8 I'd strongly recommend FR8 as stated ealier in the tread..
If the comparison was FR7 vs upgrading to FR8 then I'd stay with 7.

- Moreover ALL the versions are all just
- about EQUALLY BAD at scanning books IN THAT they deform images uglily.

This I just don't get. I get excellent OCR conversion of text scanned in 2 page
mode and bw.

On some books I have to flatten the spine (I have a ~1kg bean bag I just set on
top of the book) or it doesn't ocr correctly on the "curled" distortion at the top and bottom around
the gutter.
 
This entire thread has become too ugly to read, so I won't comment on
your message sentence by sentence any more. The purpose of this thread
is clearly stated in the first few sentences of my first message:

"I would like to add several tips in
digitizing paper documents and archiving them I noticed in course of
doing so in my personal library."

My ultimate goal is keep digital images of documents (= books and
journals in my library) as faithfully as possible, at the level of many
professional archives sites such as JSTOR and PROLA, and I wrote
several tips I thought useful for this purpose. Among them I stated
that OP15 is the best OCR program I have ever used for the purpose. I
also wrote that FR8 was much lower than my expectation and wrote a few
problems of FR6/7/8 in achieving my goal (to which you seem to have
objections anyhow).

It has become quite evident that your goal and scanning methods are
completely different from mine. If you find FR7/8 the best and OP14 a
crap for your purpose, that's OK. But your findings have nothing to do
with my findings and you can't refute my findings because of different
methodologies. Anyway, rational readers of this thread will find
whatever useful information from this thread (maybe only from my first
message).
 
Revised

This thread has become too ugly and irrational to read, so I won't
comment on your message sentence by sentence any more. The purpose of
this thread is stated in the first few sentences of my first message:
This article is a continuation of my previous article titled
"DocuMate 252/262 or fi-4120C2 for archiving documents?" I posted
on Feb 9, 2005 in this group. I would like to add several tips in
digitizing paper documents and archiving them I noticed in course of
doing so in my personal library."

My ultimate goal is keep digital images of documents (books and
journals in my library) as faithfully as possible, at the level of
professional archives sites such as JSTOR and PROLA, and I wrote
several tips I thought useful for this purpose. Among them I stated
that OP15 is the best OCR program I have ever used for the purpose. I
also wrote that FR8 was much lower than my expectation and wrote a few
problems of FR6/7/8 in achieving my goal, to which you seem to have
strong objections somehow.

It has become evident that your goal and scanning methods are
completely different from mine. You scan paperbacks casually and read
them on a reader, right? Morevover you seem to be satisfied with the
quality of dual-page scan. Such activity cannot be called "archiving"
(the subject of this and the previous threads) at any sense from the
professional viewpoint.

If you find FR7/8 the best and OP14 a crap for your own purpose, that's
OK. But your findings have nothing to do with "archiving" documents in
the above sense or you cannot refute my findings during the serious
attempt to "archive" documnents because your methodology is totally
different from mine.

Anyway rational readers of this thread will find whatever useful
information from this thread (perhaps only from my first message,
though).
 
Re-revised

This thread has become too ugly and irrational to read, so I won't
comment on your message sentence by sentence any more.

The purpose of this thread is stated in the first few sentences of my
first message:
This article is a continuation of my previous article titled
"DocuMate 252/262 or fi-4120C2 for archiving documents?" I posted
on Feb 9, 2005 in this group. I would like to add several tips in
digitizing paper documents and archiving them I noticed in course of
doing so in my personal library."

My ultimate goal is keep digital images of documents (books and
journals in my library) as faithfully as possible, at the level of
professional archives sites such as JSTOR and PROLA, and I wrote
several tips I thought useful for this purpose. Among them I stated
that OP15 is the best OCR program I have ever used for the purpose. I
also wrote that FR8 was much lower than my expectation and wrote a few
problems of FR6/7/8 in achieving my goal, to which you had a strong
objection somehow and began refuting my statements.

It is evident that your goal and scanning methods are completely
different from mine. You scan paperbacks casually and read them on a
reader, right? Morevover you seem to be satisfied with the quality of
dual-page scan and even remove headers and page numbers (ah!). Such
activities cannot be called "archiving" (the subject of this and the
previous threads) in any sense at least from the professional
viewpoint.

(Don't get me wrong. I am not saying that my method is superior and
yours inferior, or vice versa. The point should be clear.)

If you find FR7 the best and OP14 a crap for your particular purpose,
that's OK. But your findings have nothing to do with "archiving"
documents in the above sense, or you cannot refute my findings based
your own criteria and experience simply because your scan method and
purpose are totally different from mine. Should you like to discuss how
you scan double pages with a flatbed scanner and incorprate into a
reader successfully using superior FR7 and discuss how inferior OP14 is
in such a task (or whatever subject you like), this is not a right
thread. Discuss it elsewhere.

I hope that readers of this thread neglect fruitless arguments with you
and read only my first message.
 
Back
Top