Scanner Pen

INTRODUCTION , PROJECT DESCRIPTION AND WHY WE DECIDED TO BUILD IT:

The basic goal is to create an object that can recognize digital characters already printed on paper, transfer them into a computer and also modify them. 

Moreover, we would like to have an object that can transfer characters word per word in real time, so at the end of a paragraph we could have the entire editable text. 

We decided to realize this project because we would like to reduce time of transcription from paper to digital, to make instant the web (meanings, translations and so on…) and to help people who have any manual or learning difficulties.

To create an object equipped with a camera that can take a sequence of photos that will be later transferred onto a computer. The photos will be also transformed into an editable text, using a programme based on optical cha

The main goal that will satisfy us is the recognition of a single character at a time, our maximum goal is to recognize every character at once and without interruptions, while moving the pen on the characters. 

(If this thing works, we would like to add a modality that can recognize and transfer manual writing and also change it into digital writing.)

SCIENTIFIC CORPUS:

The Raspberry Pi is a sort of mini computer that plugs into a monitor and uses a standard keyboard and mouse. It is a capable little device that enables people of all ages to explore computing, and to learn how to program in languages like Python, that is an high-level programming language with dynamic semantics. 

Optical Character Recognition, or OCR, is a technology that enables you to convert different types of documents, such as scanned paper documents, PDF files or images captured by a digital camera into editable data. First, the program analyzes the structure of document image. It divides the page into elements such as blocks of texts. The lines are divided into words and then – into characters. Once the characters have been singled out, the program compares them with a set of pattern images. Basing on the hypotheses about  what this character is, the program analyzes the different variants. After processing huge number of such probabilistic hypotheses, the program finally takes the decision, presenting you the recognized text.

STRUCTURE OF THE PROJECT:

An object that has a thin, elongated shape, equipped with a camera and two supports on the extremities that have the function to keep the correct distance between the camera and the sheet so that the camera can focus on the letters. It’s necessary to have a programme based on OCR technology.

MATERIALS AND TOOLS:

Endoscope camera
small sizelow cost		8€
Raspberry pi4
more functionalities than Arduino		35€
Wooden support structure
cheapconvenience		/
Keyboard
needed		>10€
Breadboard and electricals cables
to make the circuit		6€Risultato immagini per breadboard with cables
Display
needed		/

PROGRAM:

RISK ASSESSMENT: 

During the construction of the scanner pen there were no particular dangers but some risks can be found during the processing of the wood for the support.

HOW IT WAS MADE:

  1. Learning how to convert images (JPEG) into editable texts (docx), shoot through mobile phone, use existing programs and applications (free OCR, google translate, small PDF)
  1. Purchase of an endoscopic camera to make the previous step easier
  1. Need of creating a process of converting the taken images, more immediate ⇒ using the “raspberry pi” card (flexible programming)
  1. Writing a program (mentioned above) with the help of experts
  1. Experimentation/testing of the program and creation of a circuit, for “start program” input  
  1. Creation of a support structure for the camera in order to make the photos taken as clear as possible, this facilitates the OCR transformation process

PROBLEMS AND HOW TO SOLVE THEM:

For us students of third year high school, and with little experience in information technology, facing this project was a great challenge.

None of the three participants knew the use of programming, but thanks to the help and the knowledge of the tutors and staff and teammates’ commitment we managed to create this project that made us proud.

Problems:
1. Finding a free online program that uses OCR technology

2. Finding a camera that is at the same time cheap, small size and of good quality

3. Learning the Python programming language, writing the program and combining all the steps.

We solved all the problems through online searches and thanks to help of the tutors and MIT staff

video