Book Flipping Scanning

[furtanos]

Professor of Copyism

[adelle0001]

nice one!

[unamaxify]

awesome !

now all we need is brain interface

[seglermichael]

i bet he drives a honda

[moukonsan]

韓国、中国、北朝鮮に技術盗まれないよう気を付けないとね

[easter5bunny]

是非とも商品化してほしい！

[mamatalu]

I wish I could study books like that!!

[GhostWritersForHire]

Awesome! Is it on the market yet? And how about considering converting the text to speech too?

[GamePunisher85]

I could listen to Mr. Watanabe all day. His accent relaxes me.

[0982840758]

i want bye a product, do you thing abous it, can you tell me: mobil:+84982840859, Email:sachvidan@yahoo.com.vn

[0982840758]

Mình cần mua sản phẩn công nghệ này: ai biết thông tin liên hệ mình nhé: 0982840758, Email:sachvidan@yahoo.com.vn

[TheNeoSwoosh]

K great now PLZZZZ make a video on how to build this gizmo or where to buy it I needed one for longtime

[leefuji]

YES !!!

[1990iceboy]

NO YOU ARE KIDDING MAN ?>>> LOL ITS CANNOT IN THIS SPEED

[cjnavaneet]

Liked it

[tiestomen]

owesome man!! that´s so cool & so useful!!

[panajotov]

I dont know anything more about scanning beside basic stuff, and this seems like nice idea to be worked on. AN IDEA! There is a long process from it to complete product. Often the most simple ideas can be developed into something great. At least thats how Sony made walkman. Its a schoolbook example of awesome device rated as uninteresting by test subjects in early phases. And Morita only wanted to listen some music while he was jogging or whatever :)

[timothyso]

3 observations

#1 good proof of concept

#2 good scientific inquiry

And sadly

#3 Most of the pebbles for brains comments out there will never understand #1 & #2

[KarasTGN]

Все таки японцы молодцы!

[EMIHM]

It often happens that you skip a page by flipping with your thumb so when missing a page while using this would mean that you have to check if you havent missed any pages and if you did then go and take the pic of that again. So much for high speed scanning if you have to take 5x as much time to fix it.

[HighFlyinAFGuy]

@EMIHM What are you even talking about? Try scanning a book or even a magazine with a modern day scanner. So what if you skip a page here or there and have to go back and scan it in again. Using the method in the video it's "5X" faster than current methods, and certainly wouldn't take "5X" as long. What have you created lately?

[GhostWritersForHire]

@EMIHM They should have it scan the page number to double check...it's already a scanner anyways.

[marcelosinico]

Companies that make money with artist parasitism will probably lawsuit this guy for copyright infringement. They are against technology and innovation.

[grimtango5280]

Umm its a misleading title. Its not a book flipping scanner. Its a camera that takes pictures of a book being flipped. The camera takes multiple shots of a single flip and a simple vision algorithm would determine the best page to keep. The flashing lights freeze the image to eliminate motion blur. Extremely simple. Don't see much value in it though.

[quelorepario]

@grimtango5280 haha, you speak like you know what you are talking about, yeah extremely simple, right?... although everything explained in a simplistic manner seems simple. Rocket to the moon? Yeah simple too in principle, you just need fuel and thrusters right? Man on Mars? Just need to grow your own food all the way there, protect yourself from nasty radiations and somehow have enough fuel to come back, bah pretty simple. Yeah, everything seems to be simple conceptually, the thing is doing.

[grimtango5280]

@quelorepario Well I said its extremely simple because I do shit like this everyday as a phd in electrical engineering.

[quelorepario]

@grimtango5280 since when an electrical engineer needs 3D modelling algorithms with stereoscopy?... man... if you want to bullshit at least make it credible lol

[grimtango5280]

@quelorepario Since always buddy. We MAKE this shit. You are just talking like an ignorant ass who doesn't know shit about engineering in the first place.

[quelorepario]

@grimtango5280 still wondering why an ELECTRICAL engineer would need algorithms for 3D with cameras... maybe one course yeah, but telling me that you "make" this shit everyday doesn't make sense. I hope you know the difference between electrical engineering and electronics engineering.

[diepthanhthe]

woa

[Finallythankyou]

Yeah whatever dude can you help me my math quiz.

[shoegazer666]

this is fucking stupid. there's already machines which flip the pages nicely so there's less chances of mistakes so you don't need a glorified scanner with complicated software.

[usbmouse]

disu sisutemu capucha faibu handoreddo rezorushon normaru lighto hiya?

[pHiLeverhard]

HIGH-SPEDO KAMERA FTW!

[SageSmokes]

\ / Just by dis fripping motion. Dats all! Props for the innovation!

[zingry]

AWESOME!!!



[alucardloca]

Wow, this could be soooo useful in my university library! Awesome dude!

[leblancconsultoria]

ese chino acabará con todos los libros ..... todos serán copiados faclmente

[Oblivion170]

Am I the only one that see's that the right page isn't getting scanned all the way?

[grimtango5280]

@Oblivion170 Completely agree. With an average sized book it will be impossible to eliminate book bending. Unless he utilizes multiple cameras on different angles and then use vision algorithms to stitch the images together.

[kwakwanews]

a palavra mais usada nesse video é

this.... this...

[lucasmagdiel]

Put me in a good library and give me this cool machine (fully developed, of course)... It would be nice, eh?

[MrYsosad]

@lucasmagdiel Cool idea!! But ofcourse thouse librarian guys will be like this :

No, sir, you cant scan all our books to your laptop, that we store in this huge building. >:0

[cageavvi]

wth guys?. cant you get it?. yes the machine is amazing, but go with the side of the authors and you'll see what i mean, japan, only knows how to copy, copy, copy, pirate things, fck off mothercking japan

[Scriptofmagic]

Amazing Accent ! :D

Amazing Concept !

Amazing Project !

Amazing Asians !

Amazing 21th Century :)

[3sTheDUKE]

@Scriptofmagic Level: Asian

[TheDepreso]

#WIN

[Da1Getto]

duuuude :P

[asitodoasi7878]

ESPECTACULAR!!

[Espanis]

Congratulations!!!

Excellent project, I loved it!!

[carlosabundix]

meanwhile in japan...

[suertempila]

すごい！！！！！！！！！！

[pejeno]

Wow, eso fue increible.

[wsanriv]

Nice!!



[sordinof]

720p version is broken - audio is present, but video is black. :-(

[alencar2]

これは凄い。

Inacreadible.

=)

Vai salvar o dia de muitos bibliotecários rsrs

[TheBlackNotWhite]

I am sorry but what it is about?

[yunahasnoshame]

動画で世界を煽り立てる事により新たに日本の敵を増やすのはどう­かと思います。撮影してる外国人ジャーナリストなんてある意味産­業スパイでしかないんですから。安易に映像を撮らしては駄目です­よ。日本は技術戦争の真っ只中にいる事をお忘れなく。又,共同開­発なんて甘い言葉に騙されないようにお願いします。

[sigemasajp]

@yunahasnoshame  東大はどうかはわかりませんが、多くの研究機関の研究者は成果を­上げなければ身分も保障されない環境に追い込まれつつあるのでは­とおもいます。

残念ながら何らかの客寄せパンダを吊るさなければならない現状は­悲しむべきことだと思います。

[idsxsh0]

it's awesome

[alexandrepoltorak]

You should check Elphel network cameras.

[Esperanto2082]

fantastic.......you should patent it real fast........you japanese are real good at this......incredible......have you patented it already? if you did - sell it to Steve......but this is such a simple advice )))) i think you have thought it over a billion times....anyway thatnk you for innovations......u guys rock!

[LTSV83]

手でやるより自動でペラペラめくってくれる機械ならもっと早く電­子書籍が作れるでしょう

がんばって

[HolbergEM]

Very impressive. Banzai !;)

[kodamatic]

すごいです。感動です。

[hms84]

Otsukara Sama Deshita!!!

Very Nice ;)

[HelionDark]

very nice fast and innovative

[idyama]

これで、グーグルやアマゾン、アップルに追いつけるかもしれない­。

電子書籍の未来は、書籍データをいかに早く読み取れるかにかかっ­ている。

日本語が、早く読み取れる日を、切望しています。

[ShizyDemon]

Well that was AWESOME!!

[pa5op]

awsome idea m8 :P

[siavashrezaie]

It is 1000x1000, so it's 1Megapixels

[kazoo175]

これは凄い。

[narcomanel]

did he died?

[drmwa]

I'm surprised that I'm the first person to see this, and comment on the (primitive) similarity to the book Rainbows End, where entire libraries were "preserved" by running them through a wood chipper, scanning the shreds in real time, then reconstructing them all like puzzles in software.

[Winsane]

@drmwa it is like that but that is a stupid idea. However it could be used in a really interesting way to read what has been printed over by using different light sources to scan, since people used to erase paper all the time since it was so expensive we might learn a lot if the researcher only has to put books up to it for a few minutes to gain the scans instead of the current methods which are slow and tedious.

[kelath5555]

Awesome! Even if there isn't much application in scanning, just think of what the recognition software itself could do. I can see this having applications in artificial intelligence.

[wowlethon]

awesome job bro.

[DontFearGrim]

these guys probably haven't been outside in months

[mrpinkwtc]

How much did this thing cost to build?

-----------------

whatthechrist d0t c.m

[jankoch]

interesting project but your english makes my brain hurt

[gameboy772]

He has the nerdiest Japanese voice I've ever heard.

[Winsane]

@gameboy772 not really, thats just the accent of a Japanese person who likely hasnt been abroad.

[toranuguitars]

こらすごい

発想じたいは意外とシンプルなのだろうか

[hunterr83]

Yes, it takes pictures of the page while you flip through the pages. Then, it flattens them out. A lot faster than sticking each page onto a scanner.

[Winsane]

@hunterr83 also it would never have lost letters and words at the binding of the book

[gakiyarou]

i need to get me some norwal wight

[dron69]

FAKE!!!

[IQ20000Berta]

Great job i had the same idea years ago. But a big problem you can see at 1:34. There is also the transparent backside in the scan. You must correct this with software. This could be in some cases very difficult. Good luck to the japan scientists.

[BoredInfidel]

Software or proper lighting.

[poposkidimitar]

Anyway, great work jsut want to congratulate to Takashi Nakashima, Yoshihiro Watanabe, Takashi for the smart approach and the hard work.

Thank you for sharing this with the world-even tought I have seen a different video couple of month ago using the same system (but on Japanese language).

[poposkidimitar]

Great book digitization approach! BRAVO!

Abbyy finereader really does a good OCR with only 72dpi which means that this method is completely OCR-able.

But, I would like to see another book done this way (hard-cover book with less spaces Komurobetween the lines). The book used on this video has bigger characters and the lines are double spaced...

[steviejacko]

look everyone i dont know what the machine is called all i know is you put up to ten hard back covers in and press the button and in 1 second you have captured all 10 books you then take out your usb stick and plug it inyour pc and you can access it all and it is clear as this text every single page !!

[tehdog22]

errr... yes of course

[PigsCanFly99]

500 fps camera 250 ( x 2 pages) = 1 second capture time .. states 1 minute capture time @ 0.53 .. boo boo

[ElPandaDeLaAccion]

Johnny 5 is still faster!

[Glicholas]

@ElPandaDeLaAccion and HE IS ALIVE!!

[tnoyca]

This is a rather cool approach to the time problem with digitization of books.

They'll need to significantly increase the resolution of the capture before it becomes a viable solution. 1k pixels is nowhere near high enough to get reasonable OCR accuracy.

My job deals with the digitizing of books, we've found that 5k pixels on anything A4 or larger sees a significant drop-off of OCR accuracy.

[botty75]

But the resolution he mentions isn't 1M pixel? Or am I missing something? (take into account I'm no expert... don't shoot! ^_^)

[tnoyca]

@botty75

My mistake; I was referring to the pixel density along one axis rather than the total pixels. It is indeed a 1 Megapixel sensor used in this video.

A 1k x 1k pixel sensor with an A4 page would yield a maximum scan resolution of around 85ppi. A 'good' scan would need to be at least 300ppi.

[botty75]

@tnoyca

Ok! Thanks! :)

[siokaos]

@tnoyca It's 1,000x1,000, are you sure OCR requires 5000x5000? or are you mistaken.

[tnoyca]

@siokaos

It depends on the size of the book and font size. To get good OCR results with a 'typical' font, you should have a resolution of at least 300ppi. For example, a pixel width of 1000 pixels for an image of a "typical" book (2xA5) would give you 85ppi MAX. The OCR results would be useless for full-text searching.

There are robotic book scanners on the market that will scan at 2000+ pages/hour AND give you 600ppi. This is an excellent proof-of-concept, but far from being practical.

[PindzMedia]

thats pretty cool dude, good work japan dudes

[SuperMotorfan]

cool and creative

[PodeCoet]

Coolest. Accent. EVER.

Fantastic project!

[mawo006]

I think there may be "normal" flashlight to illuminate the pages and separate "scan" from "flip". UV light may than be used for highlighting the black letters on white pages so the computer gets information on the actual curved shape of the pages. This can be combined with a camera that allows collection of visible and UV light.

[au33987]

There are thousands of books published between 1850 and 1970 that contain acidic paper made from wood pulp and are deteriorating rapidly. Trying to copy all of them before they completely disintegrate is a daunting challenge, perhaps this will help.

[11moonelf]

@ au33987: Google Books needs this one for scanning all public domain & other free books they are adding to the online collection. Wondering if it's also detailed enough to scan page numbers - in case any are missed it would then be easier to go back and find them.

[CalvinoMovies]

Amazing technology guys, just a little advice, make a more dynamic presentation to make your video more attractive, I wouldn't have watched it if I had not seen it on an article :)

5 stars anyway

[fiedel]

Impressive work.

I did not fully understand the shape-recognition part of the show. Wouldn't the ringshaped flashlights have to project some sort of pattern? But Yoshihiro san says it's "just normal lightsource". I guess thats one main ingredient of the secret sauce ... or just lost in translation?

[cjeam9199]

@fiedel The way I understood it is the doughnut lights are simply to light the page, then on the alternate frames the laser shines the series of lines on the pages as seen at 0:10  As the computer knows these lines should be a certain length, straight and evenly spaced it can model the distorted shape of the page by how these lines are distorted, and hence correct the image. Thoughts?

[Bugersaler]

voun sousand on voun sousand

[CandyLand2oh9]

Wow, like seriously, Japanese people can invent the craziest things imaginable. like the robot who can play the violin.

[bati555]

I think it's partly because technological studies are far more widespread and more immersive in their colleges.

[CandyLand2oh9]

That's probably true, but one day, we'll get taken over by robots ! D;

[roozbehn]

wow this is fucking impressinve.

imaging scanning a 350 page book in 10 seconds. still has more work to do but very very promissing. I see a lots of potential in this.

[Mrdestoryer1]

These racist pieces of crap dont realize that this is how the Japanese oppress the ones that made racist comments. cause they themselves are not smart enough to invent anything themselves so they poke and make fun of the smarter ones. well hes inventing the racist pieces of crap are pawns easily manipulated to work for them like work horse. sheep waiting to be slaughtered. dumb asses!

[TrojanDude26]

great job!

[goodday861]

nice and good

[YerrrTrax]

1:03 Normal rite Lmao!

[twentydashone]

@YerrrTrax an dees arite an dees arite eez switching!!!!!

[blarbos]

the only problem i see with this is that when people flip pages they tend to mess up sometimes and clump pages together, there should be like a mechanical flipper or something to go with it

[slashd]

@blarbos

Yeah me too, I hope the computer recognizes the missing numbers (67-68-71 *beep beep* 69-70 is missing please scan again) to fix the problem

[LordAsiram]

5* for engrish

[tuner562]

They can make cool cars, huge tvs and tiny cell phones but when it comes to the letter "L" they still fail

[kimkimnii]

@tuner562

Year..

He knows English will be worthless.

He should learn Chinese.

[hlepless]

jetvision can already do that

[TheMackAroni]

wtf??

[gymer15]

this is honorable to asians...thats the fuck

[Scklyb]

are there plans for properly timing for a flip mechanism? With timing not properly sequenced for consistent imaging there is really nothing special here.

[2bornot2b1984]

The computer reconstructs the image based on the shape of the page.

I would say being able to flip the pages of a book that fast, then have the computer compile a 2 dimensional representation of the pages and more importantly word searchable database based on instantaneous optical character recognition is something special.

Have fun scanning those pages in one by one with the current technology.

[OhSnapitsJuzDin]

I wanna stick my face in there!

and make a lottaaa faces :DDD

hahaa

[ferociousfrankie]

Japan FTW

[LUXBeaker]

@ferociousfrankie no this asian dude is wasting his time with this project. Other people have already made equipment to scan books why does this kid need to waste his time designing another one?

[slashd]

@LUXBeaker maybe those other people don't want to share their technology and knowledge for free?

[ericgrau]

Inpuuuuut

[adamx20]

Wow .. book piracy is now as easy as flipping the pages. When can it install the information into little droids that tell my brain what to think?

[jacobts]

exactly what i was thinking.

[multipurpose101]

The AAP's gonna be all over this like the RIAA on Limewire and the MPAA on ThePirateBay

[zekls1]

what does it do?

[Winsucker]

that was epic

[flashfith]

my comments aren't showing wtfffff

[nemaat]

Very nice. Interesting future coming soon.

[grawlier]

This is kinda pointless o_O

[xaniared]

GREAT JOB

[cancerish]

Japanese... only when you think it can't go any weirder...

[DanReinhold]

johnny 5 is pissed

[Eddyisgreat]

Great work over in Tokyo as usual. Now sell it to Fujitsu so they can commercialize it. You'll be rich, I'll have one, and I can give my books away (while keeping digital copies). Everyone wins!!

[Jesperhkr]

cool !!! hope he gets rich of it! :D

[rough77]

but their girls got "hair" and their guys don't talk 'bout 22 cm, they talk about 11 cm when they want to impress...

[IceKing2626]

and.......what are u going to do with a camera that read books?

[purpletaco542]

woooow

[kennedy1134]

haha if you do the transribe thing when the japanese guy is talking it comes out with random stuff! haha

[The2DUDESsweet]

likaaaa sooombooody!

[Tehsty]

i dont get it

[reinventnv]

FUKIN aMAZING

[georgewig]

It's simple, just get a high fps camera and slow it down so you can see whats on the page, there's plenty of video on this type of thing here on youtube.

[mastershake1000]

Holy sheesh that's awesome!!!! Ebooks to the max!

[footballstar227]

JAPZ DID IT AGAIN NEXT IS GIANT ROBOTS

[purpletaco542]

@footballstar227 done

[sirboyking]

EPIC!

[imxlnt2]

people are just trying a new way of burning books without actually burning the book lol. technology will soon fail!

[RaVeN1K]

no they wont...