Superfast Scanner Lets You Digitize Book By Flipping Pages

Tokyo University researchers develop scanner that can capture 200 pages in one minute

2 min read

Scanning a book manually, page by page, is a slow, mind-numbing task. Google has had some ideas on how to speed up the process, but apparently these methods are still limited by how fast human hands can manipulate a book's pages before an image-capturing system.

Google, meet Masatoshi Ishikawa, a professor at the University of Tokyo. Ishikawa is well known in robotics circles for his Matrix bullet time-style amazing demos—like a robot hand that can catch objects in midair with superhuman speed. How does it work? He built a “Super Vision Chip” (that's what he calls it) that can “see” events too fast for the eye.

Ishikawa and his colleagues are already working on several applications—including a microscope that can track individual bacteria and a video game motion-capture system for gesture playing. Late last year when I visited the lab, they showed me their latest creation: a superfast book scanner.

The system, developed by lab members Takashi Nakashima and Yoshihiro Watanabe, lets you scan a book by rapidly flipping its pages in front of a high-speed camera. They call this method book flipping scanning. They told me they can digitize a 200-page book in one minute, and hope to make that even faster.

The camera operates at 500 frames per second, with a resolution of 1280 by 1024 pixels. For each frame, the system alternates between two capture modes. First it shines regular light on the page and captures text and images. Then a laser device projects lines on the page and the camera captures that as well.

The scanned pages are curved and distorted, but the researchers found a way to fix that. The laser pattern allows the system to obtain a page's three-dimensional deformation using active stereo methods. So they wrote software that builds a 3D model of the page and reconstructs it into a regular, flat shape.

The system is currently a prototype that occupies an entire lab bench. But in the future, they hope to simplify and miniaturize it for integration into portable devices like a smartphone. So one day you might be able to flip the pages of a book in front of your iPhone and get a digitized version in seconds.

True, it might not be a perfect, high-resolution copy of the book. The scanning process might skip pages and, well, your fingers might appear in the images. And that's not to mention all the copyright questions.

In fact, Watanabe told me he was particularly interested in scanning manga comics. Imagine, he said, if all of Japan's vast manga archives, at libraries, homes, and elsewhere, could be rapidly scanned and shared among manga fans around the world. That'd be nice. Alas, when he contacted one publisher, they didn't like his idea and forbade him from using their books for testing the scanning device. Watanabe currently uses a mock book he made himself.

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman
LightGreen

This article is part of our special report on AI, “The Great AI Reckoning.

"I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓ Show less