Project

# Title Team Members TA Documents Sponsor
76 Tool that translates printed text to braille
Abraham Han
Blas Alejandro Calatayud Cerezo
Samuel Foley
Raman Singh design_document1.pdf
final_paper1.pdf
photo1.jpg
photo2.jpg
presentation1.pdf
proposal1.pdf
video
# **Tool that translates printed text to braille**

Team Members:
- Blas Alejandro Calatayud Cerezo (bac10)
- Abraham Seungyeop Han (shan79)
- Samuel Foley (safoley2)

# **Problem**

According to the World Health Organization, currently there are around 39 million people who are legally blind around the world. Right now there are not many resources available for people who can only read braille to read physical written text from a book or magazine, and those that are available are very expensive.

# **Solution**

Our solution is to create a tool that can be placed over printed text and translate it to braille so that blind people can read it. This tool will be divided into two parts that will be connected between each other through several wires that will transmit power and data.

The first part will be a handheld device with a camera to recognize the letters in a word. The user would hold this handheld device with one hand and place it on top of written words.
The second part will be a box that will contain the pcb with the microprocessor and an external battery module. It will receive the images taken by the camera, process them to recognize every letter on the word and finally output on top in braille the characters of that word one by one using pins that can be pushed up and down to create braille characters.

The person using this device will place one of their fingers on top of the moving pins used to create the braille characters to read the printed text.
After showing all the braille characters in a word, the user can simply move to the next word for it to be shown in braille.

# Solution Components

# Subsystem 1 - Handheld Housing

The handheld housing will have the camera sensor attached to it, which would be transmitting image data to the microcontroller. The housing will have to be ergonomic to hold and made of some lightweight material, like plastic. We may additionally add some way for the user to attach the housing (e.g. velcro straps) for convenience.

# Subsystem 2 - PCB containing the MicroController

A custom PCB will be designed in order to connect all other subsystems. The PCB would connect the pin motors for the braille “display”, the handheld housing containing the camera sensor, and the external battery module in order to power all the other components. The PCB would also control the recharging of the battery module to ensure optimal battery health.

The microcontroller will take images from the camera sensor to process the text characters in the image. The image processing, or more specifically the OCR (Optical Character Recognition), will be done through open source computer vision and machine learning libraries such as OpenCV or Tesseract. The microcontroller will also control the motors that will drive our pins to form braille characters.

Also, a Raspberry Pi or a similar microcontroller will interface with the microcontroller used in our custom PCB as a more powerful chip may be required for better OCR performance. However, Arduino could also be a viable option.

# Subsystem 3 - External battery module

An external battery module will power the whole system and:
1. Be lightweight for ease of transporting
2. Powerful enough to sufficiently power the whole system
3. Have theoretical “all-day” battery-life

The battery module will be made up from LiPo (Lithium Polymer) batteries for their high energy-density. Potentially multiple batteries hooked up together for a battery array depending on the energy needs. Such a module would need to have a casing of some sort (simple plastic casing would suffice) in order to protect the batteries from the outside environment.

Some downsides to such a battery module would be that LiPo batteries require extra care in their recharge cycles as they must be evenly charged, and also not overcharged. Further, LiPo batteries can become hazardous if punctured, and such a safety hazard would have to be addressed through the design of the casing.

# Subsystem 4 - Motors for Pins

Linear actuators controlled by the microcontroller will be used to move up and down 6 small bars through holes made on top of the box to form braille characters. The bars required to form each character in braille will move up and down in a synchronous way so that the user can read them with their finger.

# Criterion for Success
- The moving handheld camera can take pictures of every letter in a word and send them to the microcontroller.
- The microcontroller can recognize every letter in a word using the images sent by the camera.
- The microcontroller can translate the recognized letters into a series of braille characters for the pins to make.
- Linear actuators can push pins in a synchronous way to create braille characters that are easy to read.

Healthy Chair

Ryan Chen, Alan Tokarsky, Tod Wang

Healthy Chair

Featured Project

Team Members:

- Wang Qiuyu (qiuyuw2)

- Ryan Chen (ryanc6)

- Alan Torkarsky(alanmt2)

## Problem

The majority of the population sits for most of the day, whether it’s students doing homework or

employees working at a desk. In particular, during the Covid era where many people are either

working at home or quarantining for long periods of time, they tend to work out less and sit

longer, making it more likely for people to result in obesity, hemorrhoids, and even heart

diseases. In addition, sitting too long is detrimental to one’s bottom and urinary tract, and can

result in urinary urgency, and poor sitting posture can lead to reduced blood circulation, joint

and muscle pain, and other health-related issues.

## Solution

Our team is proposing a project to develop a healthy chair that aims at addressing the problems

mentioned above by reminding people if they have been sitting for too long, using a fan to cool

off the chair, and making people aware of their unhealthy leaning posture.

1. It uses thin film pressure sensors under the chair’s seat to detect the presence of a user,

and pressure sensors on the chair’s back to detect the leaning posture of the user.

2. It uses a temperature sensor under the chair’s seat, and if the seat’s temperature goes

beyond a set temperature threshold, a fan below will be turned on by the microcontroller.

3. It utilizes an LCD display with programmable user interface. The user is able to input the

duration of time the chair will alert the user.

4. It uses a voice module to remind the user if he or she has been sitting for too long. The

sitting time is inputted by the user and tracked by the microcontroller.

5. Utilize only a voice chip instead of the existing speech module to construct our own

voice module.

6. The "smart" chair is able to analyze the situation that the chair surface temperature

exceeds a certain temperature within 24 hours and warns the user about it.

## Solution Components

## Signal Acquisition Subsystem

The signal acquisition subsystem is composed of multiple pressure sensors and a temperature

sensor. This subsystem provides all the input signals (pressure exerted on the bottom and the

back of the chair, as well as the chair’s temperature) that go into the microcontroller. We will be

using RP-C18.3-ST thin film pressure sensors and MLX90614-DCC non-contact IR temperature

sensor.

## Microcontroller Subsystem

In order to achieve seamless data transfer and have enough IO for all the sensors we will use

two ATMEGA88A-PU microcontrollers. One microcontroller is used to take the inputs and

serves as the master, and the second one controls the outputs and acts as the slave. We will

use I2C communication to let the two microcontrollers talk to each other. The microcontrollers

will also be programmed with the ch340g usb to ttl converter. They will be programmed outside

the board and placed into it to avoid over cluttering the PCB with extra circuits.

The microcontroller will be in charge of processing the data that it receives from all input

sensors: pressure and temperature. Once it determines that there is a person sitting on it we

can use the internal clock to begin tracking how long they have been sitting. The clock will also

be used to determine if the person has stood up for a break. The microcontroller will also use

the readings from the temperature sensor to determine if the chair has been overheating to turn

on the fans if necessary. A speaker will tell the user to get up and stretch for a while when they

have been sitting for too long. We will use the speech module to create speech through the

speaker to inform the user of their lengthy sitting duration.

The microcontroller will also be able to relay data about the posture to the led screen for the

user. When it’s detected that the user is leaning against the chair improperly for too long from

the thin film pressure sensors on the chair back, we will flash the corresponding LEDs to notify

the user of their unhealthy sitting posture.

## Implementation Subsystem

The implementation subsystem can be further broken down into three modules: the fan module,

the speech module, and the LCD module. This subsystem includes all the outputs controlled by

the microcontroller. We will be using a MF40100V2-1000U-A99 fan for the fan module,

ISD4002-240PY voice record chip for the speech module, and Adafruit 1.54" 240x240 Wide

Angle TFT LCD Display with MicroSD - ST7789 LCD display for the OLED.

## Power Subsystem

The power subsystem converts 120V AC voltage to a lower DC voltage. Since most of the input

and output sensors, as well as the ATMEGA88A-PU microcontroller operate under a DC voltage

of around or less than 5V, we will be implementing the power subsystem that can switch

between a battery and normal power from the wall.

## Criteria for Success

-The thin film pressure sensors on the bottom of the chair are able to detect the pressure of a

human sitting on the chair

-The temperature sensor is able to detect an increase in temperature and turns the fan as

temperature goes beyond our set threshold temperature. After the temperature decreases

below the threshold, the fan is able to be turned off by the microcontroller

-The thin film pressure sensors on the back of the chair are able to detect unhealthy sitting

posture

-The outputs of the implementation subsystem including the speech, fan, and LCD modules are

able to function as described above and inform the user correctly

## Envision of Final Demo

Our final demo of the healthy chair project is an office chair with grids. The office chair’s back

holds several other pressure sensors to detect the person’s leaning posture. The pressure and

temperature sensors are located under the office chair. After receiving input time from the user,

the healthy chair is able to warn the user if he has been sitting for too long by alerting him from

the speech module. The fan below the chair’s seat is able to turn on after the chair seat’s

temperature goes beyond a set threshold temperature. The LCD displays which sensors are

activated and it also receives the user’s time input.

Project Videos