Showing posts with label OCR. Show all posts
Showing posts with label OCR. Show all posts

Friday, February 06, 2009

The library is now open: Google Books Mobile employs OCR and adds unintended accessibility

This latest bit of news is, well, hot off the press.

In
Google’s
on-going quest to make the world a more searchable place, and to make all discovered content available for viewing on whatever device one is using, whether that is a computer or mobile phone, the internet search giant has learned to embrace a tool of the assistive technology trade, namely optical character recognition, or OCR.

This all came about in Google Books latest quest to put
1.5 million books in your pocket.

One of the great things about an iPhone or Android phone is being able to play Pacman while stuck in line at the post office. Sometimes though, we yearn for something more than just playing games or watching videos.

What if you could also access literature's greatest works, such as Emma and The Jungle Book, right from your phone? Or, some of the more obscure gems suchas Mark Twain's hilarious travelogue, Roughing It? Today we are excited to announce the launch of a mobile version of Google Book Search, opening up over 1.5 million mobile public domain books in the US (and over half a million outside the US) for you to browse while buying your postage.

While these books were already available on Google Book Search, these new mobile editions are optimized to be read on a small screen. To try it out and start reading, open up your web browser in your iphone or Android phone and go to
http://books.google.com/m.



What the fine folks at Google found when taking this endeavor mobile was that the page images Google Books serves up as pictures of individual pages on a computer are quite “unwieldy” on mobile phones. To solve the dilemma, they have employed OCR to extract the text of the book pages and display this content just as it would show any other web page.

And, in the above post, Google also clearly illustrates some of the inherent imperfections blind OCR users have known about for years. Yes, some times the text comes out as gibberish. It isn’t perfect, but it works more often than not.

So, here’s the really big news in all this. Even though it was an unintended result, Google Books has finally made its material accessible to blind screen reader users. And the best news in all this is that these folks don’t even have to be using the web on a mobile device with a screen reader. All they need to do is access the
Google Books Mobile site

I just tried it on my pc and it works great. I searched for Mark Twain and clicked on
A Connecticut Yankee in King Arthur’s Court.

While it took some time for Google to discover the added value of universal design, perhaps they can now begin to see it for the benefit that it truly is. If they had just used this OCR presentation method from the inception of Google Books, the “Blind Folks Not Allowed” sign wouldn’t have been necessary.

So, even though it was an unintended result, hat’s off to Google for finally bringing accessibility to Google Books. Yes, the library is now open.

Saturday, May 31, 2008

Interesting, intriguing interview with Ray Kurzweil

It is early on Saturday morning and I’m buzzing. No, it’s not due to the second cup of coffee I’m enjoying right now. I awoke around 3 a.m. and tried without success to go back to sleep. Resigned to staying awake for a while, I grabbed the TV remote and began flipping through programs. There were a number of news talk shows airing the latest flap in the Democratic Presidential race and I finally found an episode of The Brady Bunch to dull my senses, hoping to relax me back to sleep. However, when that show ended, it was 4 a.m. and I began surfing channels again.

I then stumbled upon a program that instantly energized me. When I heard the host say that he had been trying to get his next guest on for the last ten years, I became curious. But, when he announced his guest’s name, I knew I was going to be up for awhile and was beyond going back to sleep. I got up and put on the coffee so that I could take it all in.

I’ll bet that my ears visibly jumped. I was immediately pumped up when I heard that Ray Kurzweil was the guest on the
Glenn Beck show
on CNN Headline News. This was a repeat of the show's original broadcast from the night before, Friday, May 30, 2008.

I don’t know if CNN airs entire one-hour shows such as Beck’s online, but if they don’t, they should make an exception for this one. It is a riveting exchange with this fascinating man. However, if you missed the program and cannot see it in any re-airing, you can
Read the show’s transcripts here.

If you are not familiar with his name, Kurzweil is an author, inventor, futurist, and man of many other appropriate titles, but most of all, I believe that he’s one of the smartest men of our time. His name is synonymous with assistive technology. One of the first pieces of AT that I purchased was a Kurzweil Reading Edge, a stand alone optical character recognition (OCR) scanner that sold for more than $5,000 in late 1994. I remember that being the year when I made that purchase, because I made it shortly after leaving the Criss Cole Rehabilitation Center, which is where I first learned about that innovative machine.

Kurzweil is in high demand as a speaker at numerous conferences and events whose subjects range on a variety of issues from music to science. While the conferences might vary, his focus always centers on information technology. If you didn’t catch it in my recent post, He was one of the two noted keynote speakers at this year’s CSUN.

I mentioned that I had purchased my Reading Edge in 1994. That might have sounded like cutting edge technology for that year, but that invention was actually already two decades old at that time. Kurzweil invented the world’s first character recognition machine that was the size of a washing machine and sold the first model in 1976. Watch the interview or read that transcript to hear how that whole idea evolved. I think you’ll find it interesting.

(Extra credit if you can guess the name of the first customer of this product. Read to the end of this post for the answer.)

Later in the Beck show, Kurzweil demonstrated the latest incarnation of his reading machine, the
KNFB Reader.
He scans a document and lets the product do its thing, reading on the show for the entire world to see and hear. Kurzweil does promote the KNFB Reader for its dual purposes of assisting people who are both blind/visually impaired and dyslexic. To emphasize this aspect, he shows the phone’s screen to demonstrate how the words being read from the scanned document are highlighted, and how the highlights move as the speech progresses.

Thing about what that says. From a behemoth, washing machine model to a four-ounce cell phone that can be held in the palm of your hand in just over 30 years…Wow!

To understand the evolution of that OCR system is to understand most of what Kurzweil discusses. Exponential growth is key to understanding how things evolve in our world. Kurzweil is a highly intelligent man, but in this interview, he doesn’t come across as an overly intellectual person speaking in language that the layperson can’t understand. With Beck providing the questions, Kurzweil explains his understanding of life and technology in very understandable terms.

Don’t even get the idea that Kurzweil is one-dimensional. Kurzweil speaks of many things during this program and assistive technology is just one of them. I mentioned that he is a futurist. Don’t confuse that with some flaky person who says they are a fortuneteller. Futurists examine history, technology, and trends on several planes and make predictions from the understanding gleaned from that examination. Kurzweil’s been doing this for thirty years and has been pretty accurate on many fronts.

Here are a few points from this interview I find totally intriguing.

* Exponential growth is based on numbers doubling and the time needed to do this.

* The energy produced by the sun is more than 10,000 times to power needed to support our world today.

* Solar power technology is doubling every two years.

* We are only five years from solar power being more affordable than current coal or gas.

* We are only seven doublings away from solar energy being the dominant power supply for our earth.

If you can find it, do check out the interview. You’ll discover that there is a good discussion of health and the integration of technology with our bodies, as well as a good delving into the subjects of fossil fuels, computers, information storage, global economics, and even briefly touches on the applications of information technology as it applies to terrorism. You'll also learn a little bit about Kurzweil himself.

And, here’s your extra credit answer: Kurzweil sold his first character recognition machine to Stevie Wonder. The two forged a friendship from this that led to further collaboration. The offspring of this work was the invention of the world’s first electronic piano.

By the way, my coffee’s now all gone and I’m still buzzing.

Monday, April 07, 2008

ScanR.com offers potential for free, portable OCR

I’m not certain how well this would work, but there is a web-based application that may provide the same functionality to blind people that the KNFB Reader Mobile provides. And, if it works like it supposed to, portable optical character recognition (OCR) can be had for free!

ScanR.Com
Is an innovative tool that allows a person to take a picture with their cell phone camera, send it to ScanR, and, within seconds, receive an email with an attached pdf document with OCR performed on the image. (ScanR uses the ABBYY OCR software, so if you’re familiar with that OCR engine, then you have an idea on what to expect as far as the quality of the OCR work.)

Because this service employs some of the common tools available on today’s smart phones, which are a requirement to run either of the commercially available mobile screen readers, this seems like an ideal fit for blind people who use these phones. It employs digital photography, email, and the reading of PDF documents.

ScanR.com is a subscription service, but does have a limited free service, allowing a user to submit five images a month at no cost. Free documents are returned with the ScanR logo on them. The logo is removed when subscribing to the subscription program. After all, while the fee costs money, it is still very reasonable-- only $3 per month or $30 a year.

I’m interested to know how well this works for blind people. Perhaps it would be the ideal solution for a restaurant menu, or a handout at a meeting. Being that I don’t have a data plan with my mobile phone service, I’m not in a position to personally try this. If you or somebody you know does and you’re willing to try this, I’d love to know how well the service works. If so, please leave a comment.

If it works as well as it is presented, when compared to the cost of the latest model of the KNFB Reader on a Nokia N95, with its $2000+ price tag, the fee-based ScanR offers a significant value.

Additional reading:
I originally learned about this service through an article in the
New York Times.
There are also a couple of good write-ups from last year in the
Wall Street Journal
And
Newsweek.

Monday, January 28, 2008

Updated: New KNFB Reader promises another advance in portable OCR abilities

**See end of this post for an available audio interview.

In October, 2006, I finally got the chance to get my hands on a KNFB Reader and take it for a test drive. The KNFB Reader is the breakthrough in portable OCR scanning that unhooked that assistive technology task from a computer. It joined a PDA running Kurzweil OCR software with a digital camera to input the graphical image as the document being scanned. This concept was landmark in that it allowed blind and visually impaired people the ability to be anywhere and scan work documents, restaurant menus, and any other document with text in standard fonts.

At that time, while I marveled at the concept of having the ability to virtually scan any document at any location, I was also struck with both the unit’s somewhat slow speed and bulkiness. I felt that the processing speed should have been quicker and also, because it was a camera paired with a personal data assistant, it was a little bulky to make it what I felt was truly portable.

It seems like I wasn’t the only one who felt this way, at least about the size concern. The latest evolution of this product is here. The new
KNFB Reader
Is about to be released and it runs on a Nokia N82 mobile phone with a built-in digital camera.

Now, that really makes some sense. Today’s smart phones are virtually running a small computer already to execute all the semi-computing tasks anyhow, which the market demands, and this is what the latest incarnation of the tool is using to run the Kurzweil software. And, of course, the camera was already built-in, so there is no need to use two digital devices to do one job.

According to the press release linked above, which is basically just announcing today’s press conference of the KNFB Reader’s advanced enhancements, the camera is implied to also be faster than its predecessor. At least that’s what I take from the words of the President of the National Federation of the Blind (the NFB in KNFB).

And, According to the below paragraph (also from the press release,) which outlines the accessibility features of the software, I’m also going to take it that the program works as a screen reader for the phone.

“Blind users will have access to all of the functions featured in the most advanced cell phones on the market including video and music playback, GPS, wireless communications, photography, e-mail, text messaging, calendar and task functions, and more. The combination Reader and cell phone weighs 4.2 ounces and can store thousands of printed pages with easily obtainable extra memory. Users can transfer files to computers or Braille notetakers in seconds.”

This all sounds wonderful, but I want to know what this will cost and when it will be available. I’ve check both the press release and the official
KNFB Reader web site,
For more information. However, I have not found any specifics about when this dreamy piece of assistive technology will be available, nor what the price will be.

Hopefully, the unit’s price will not be stratospheric ($3,500) like its predecessor. This is especially applicable when the
Mobile Speaks screen reader
For smart phones, including two Nokia models, Is available commercially through AT&T for only $89. Yes, the AT&T offering doesn't include the OCR capabilities, but it gives everything else. Given that blind consumers have something to compare it to, that makes the price tag for the OCR software stand out and may make some wonder how much that is truly worth. If the access to the phone's features can be had for only $89, that makes it imperative for the KNFB Reader to be priced somewhat competetively.

I feel certain more will be revealed in coming days. Check back here soon for updates.

***
Update 01-28-08 7:15 pm

Okay, here is the updated information on pricing and shipping.

Celia Black, Director Marketing Communications for the KNFB Reader, read my post and sent me the following information for the new KNFB Reader.

“The retail price of the knfbReader Mobile software is $1595 US. The NOKIA N82 sold separately, ranges between $530 and $600.”

She provided the following web sources to check for pricing the phones:

http://www.egizmo.com/servlet/the-910/Nokia-N82-5.0-MP/Detail

http://www.mobilecityonline.com/wireless/store/productdetail.asp?productid=22925&refid=GG-N82&gclid=CNztrJGDmpECFQI2sgod5mfFOw

She also informs me that the new KNFB Reader will begin shipping on Feb. 15, 2008. There is
authorized dealer information
on the KNFB web site.

Additionally, according to
The Ranger Station
one vendor is selling an all set up, ready to use mobile phone with the KNFB software and either the TALKS or Mobile Speaks screen reader, which will run a cool $2,495.

***
Update 02/01/08

The Blind Bargains web site
has an audio interview available, with James Gashel, Vice President of marketing for KNFB Reading Technologies. Mr. Gashel discusses the changes that have been made, as well as what may lie ahead for future versions of the technology.