photo pdf or clearscanned pdf can be processed quicker?

[taylor3456]

I have an Eink reader, Hanvon N800, with 8inch screen. But the cpu of it is not very fast, only 360MHz.

I mainly use this device for reading pdf files. I have some scanned pdf files. But I do not know if I should clearscan(an option in acrobat ocr) it. From intuition, "the searchable image(exact)" (an option in acrobat ocr) should be slow for ereaders to process, perhaps also for "the searchable image".

But I have no clue, if the original photo pdf file(before ocr) or clearscanned pdf file(after ocr) should be processed(quickly turning pages) quicker?

I hope you can understand what I mean. Please feel free to say some thing about your experience and opinions!

[DDHarriman]

Hello

You are right, clearscan has the lower impact.
Basically clearscan gives the most proximal result of if you would have writen and formatted the text/images of the pdf yourself.
The catch is that all the errors (incorrect or not recognized characters/word/phrases) will show, and one must correct them by hand - Acrobat is not the best tool to do this correcting - “proof reading” is the term for it.

The other two options are what one calls a “two layer” pdf: one layer is the original (or compressed) image and the other the text (the result of the ocr processing), thus occupying the size and putting (at least) the same pressure in the eBook reader as if you were just reading a pdf made from the scanned images.
In practice, for what your problem, doing an searchable image ocr (exact or not) on Acrobat is useless.

Best regards,

[taylor3456]

Originally Posted by DDHarriman

Hello

You are right, clearscan has the lower impact.
Basically clearscan gives the most proximal result of if you would have writen and formatted the text/images of the pdf yourself.
The catch is that all the errors (incorrect or not recognized characters/word/phrases) will show, and one must correct them by hand - Acrobat is not the best tool to do this correcting - “proof reading” is the term for it.

The other two options are what one calls a “two layer” pdf: one layer is the original (or compressed) image and the other the text (the result of the ocr processing), thus occupying the size and putting (at least) the same pressure in the eBook reader as if you were just reading a pdf made from the scanned images.
In practice, for what your problem, doing an searchable image ocr (exact or not) on Acrobat is useless.

Best regards,

thank you!

I usually keep the results of clearscanned pdf and do not do the "proof reading" because the unrecognized character/word/image do not bother me very much.But I have the same feeling with you about the page turning speed with the searchable images(with or without exact)---these files should lag the page turning speed of the eink devices due to the big size!

Acctually I wanna share some results with you and other pdf readers:

I tested 50 pages of the originally same pdf document. I will turn pages 50 times on the pre-clearscanned one and after-clearscanned one.

When the pages contain many graphs or images, the reader turn pages faster with the pre-clearscanned one.
When the pages do not contain many graphs or images, there is no big difference between 2 files.

Sometimes the reader will lose some characters or part of the graph when turning pages with the after-clearscanned one.

[DDHarriman]

Hi

Sorry to come back to you just today, but I have been not in the forum the all week.

Your experience is correct, it’s also my experience in all my eBook readers.

I do not know the hanvon eBook reader, but one of my older ones, the Cybook Gen, must have a less power processor then yours, and also the rendering of pdf files in eBook reader is related with the operating system and the program that reads the files.

Anyway, if the program that reads the pdf encounters a complicated pdf page - complicated formatted text, images and graphics -, it has to “work” more to show the resulting page, than just to show an image (witch is in reality a not ocred pdf page).

The better experience and speed on rendering pages and “flipping” pages one gets is to convert (ocr is one of the steps on the process) the original images (or image pdf files) into the main reading format the eBook reader is designed for (epub, mobipocket, or any other).

Best regards,