Report
Report
Report
Generally dumb people use sign language for communication but they find difficulty in
communicating with others who don’t understand sign language. This project aims to lower
this barrier in communication. It is based on the need of developing an electronic device that
can translate sign language into speech in order to make the communication take place between
the mute communities with the general public possible. A wired data gloves is used which is
normal c gloves fitted with flex sensors along the length of each finger and the thumb. Mute
people can use the gloves to perform hand gesture and it will be converted into speech so that
normal people can understand their expression. Sign language is the language used by mute
people and it is a communication skill that uses gestures instead of sound to convey meaning
simultaneously combining hand shapes, orientations and movement of the hands, arms or body
and facial expressions to express fluidly a speaker’s thoughts. Signs are used to communicate
words and sentences to audience. A gesture in a sign language is a particular movement of the
hands with a specific shape made out of them. A sign language usually provides sign for whole
words. It can also provide sign for letters to perform words that don’t have corresponding sign
in that sign language. In this project Flex Sensor plays the major role, Flex sensors are sensors
that change in resistance depending on the amount of bend on the sensor. Various sign language
systems have been formulated by manufacturers around the world but they are neither
adaptable nor cost -effective for the actual users. This report presents a system prototype that
is able to automatically recognize sign language to help normal people to communicate more
effectively with the diminished people.
5
INTRODUCTION
About nine thousand million people in the world are deaf and dumb. The communication
between a deaf and normal person is to be a serious problem compared to communication
between blind and normal visual people. Sign language is a non-verbal form of
intercommunication which is found amongst deaf people in world and is a more organized and
defined way of communication in which every word or alphabet is assigned some gesture. The
languages do not have a common origin and hence difficult to translate. A gesture in a sign
language is a particular movement of the hands with a various shape made out of fingers. A
gesture on the other hand, is a static shape of the hand orientation to show a sign. Gesture
recognition is categorized into two main groups i.e. vision based and sensor based. The sensor
based technique offers better mobility.
The main aim of this project is to present a system that can efficiently interpret Sign Language
gestures to both text and auditory speech. We have focused on designing a Human Computer
Interface (HCI) system that can understand the sign language accurately so that the signing
people may communicate with the non signing people without the need of an interpreter. It can
be used to generate speech or text. The converter here makes use of a glove based technique
consisting of flex sensors. The device translates alphabets as well as can form words using
specific gestures made by the person.
This project also deals with the design and development of a robotic hand with real-time
control, which is precise and cost-effective. This five-fingered robotic arm mimics a small
degree of dexterity for deaf & dumb. The basic components required in this project is the hand
itself, the Arduino, the glove, and the flex sensors, LCD and SDcard module. The glove is
mounted with flex sensors: variable resistors that change their value when bent. The Arduino
reads the voltage change when the sensors are bent, and triggers the servos to move a
proportional amount.
6
PRODUCT SPECIFICATIONS
HARDWARE DETAILS
1. Arduino UNO
7
2. Flex Sensor
Flex sensors are sensors that change in resistance depending how much the sensor is
bend. Sensors convert the change in bend to electrical resistance - the more the sensor
bend, the higher the resistance value. Using the Flex Sensor is very easy. There are
couple of different manufacturers in the market. Datasheet instructs you to use
operational amplifier (opamps). That may be useful if you plan to use flex sensor as
stand-alone device (without any microcontroller).Because We are using arduino, We
skipped all OpAmps and made a very simple circuit with only one additional resistor.
Varying the value of the resistor will results different readings. With 10k Ohm resistor
we will get the desired values. This works fine for us. In our code we assumed that all
values above 900 mean that the sensor is bend. All values below 800 mean that sensor
is nor bend. Note that Flex sensor give reliable readings ONLY if you bend it on the
specific direction (usually towards on the text side of the sensor).
8
3. Liquid Crystal Display (LCD)
The output of the system is displayed on LCD. In our system16x2 LCD is used. The
16x2 indicates 16 columns and 2 rows. So, we can write 16 characters in each line. So,
total 32 characters we can display on 16x2 LCD.The LCD will display the word
corresponding to each hand gesture made by user. This enables the mute people to
convey their thoughts using words. And also helps deaf people to understand what is
conveyed the person communicating with them. As deaf people can read the words
directly from the LCD. Hence this system enables deaf people to communicate with
Mute as well as normal people.
4. SD Card
The purpose of using SD Card in this system is to store the data. In our sys tem the
16GB SD card is used. The SD card stores the .wav file corresponding to each word.
This data used to play the output of the system on Loud Speaker and display it on the
LCD screen.
9
IMPLEMENTATION AND CONSTRUCTION
The flex sensors require a circuit in order for them to be compatible with Arduino. It's a voltage
divider: the flex sensors are variable resistors, and when paired with resistors of a static value,
change in resistance (in this case bending the sensor) can be sensed through the change in
voltage between the resistors. This can be measured by the Arduino through its analog inputs.
10
The schematic is attached (red is positive voltage, black is negative, and blue goes to the
Arduino). The resistors in the photo are 10K.
The main GND wire, which is connected to all the individual GND wires from the sensors,
gets plugged into the Arduino's GND. The positive end of each of the Flex sensor is connected
to one end of the potentiometer which is used to control the voltage provided by the sensors.
Now it's time to mount the sensors and their circuit onto the glove itself. On each finger, with
a pencil or pen, we made small lines over the tops of each joint/knuckle placed on the inside
rather than outside and pasted them on the glove using Double sided Tape.
11
WORKING:
Physical form of the circuit is shown in figure below. I consists of microcontroller interfaced
flex sensors, voice module, etc. Change in the values of flex sensors gives some hex code to
the microcontroller which after compilation displays the output in LCD and also produces the
voice through the speaker. The words or signs are obtained by taking English as a reference.
For every word, values of flex sensors are compared with the values already saved in the
microcontroller and then the result is displayed in LCD.
Flex sensor and microcontroller are used in this system to capture the words. The work of flex
sensor is to obtain changed position of fingers and to capture words.By changing the position
of hands the values of flex sensor will not get changed because flex sensor is placed in fingers.
Sign language used by deafand dumb people will be using fingers or hands in marked position
and rotating hands or fingers. To analyze these positions of fingers or hand microcontroller is
used. This microcontroller is placed on the device so that by changing the finger for
conversation, flex sensor and microcontroller values will get changed, by comparing these two
values, output is displayed in display and voice module gives the voice output.
12
SOFTWARE DESIGN:
The "Sign Language Trainer & Voice Convertor” software receives the values given by the
flex
on the two gloves through an Arduino Microcontroller Board. The software end has the
following features:
The software is based on the statistical template matching model and the entire model can be
divided into three
parts, namely: calibration of the sensors, training of the model and gesture recognition.
Calibration of the sensors is achieved by taking the minimum and maximum sensor values and
then normalizing
and quantizing the values, so as to convert the read sensor values into a pre-defined range of
discrete data set.
13
SOFTWARE ALGORITHM
#include <LiquidCrystal.h>
#include <SD.h> //include SD module library
#include <TMRpcm.h> //include speaker control library
void setup(){
void loop()
{
delay(1000);
v1=analogRead(f1);
v2=analogRead(f2);
15
v3=analogRead(f3);
v4=analogRead(f4);
v5=analogRead(f5);
// v1=map(v1,700,900,0,255);
Serial.print("Flex 1 = ");
Serial.println(v1);
Serial.print("Flex 2 = ");
Serial.println(v2);
Serial.print("Flex 3 = ");
Serial.println(v3);
Serial.print("Flex 4 = ");
Serial.println(v4);
Serial.print("Flex 5 = ");
Serial.println(v5);
delay(100);
if (v1<800)
a1=0;
else if (v1>900)
a1=1;
if (v2<900)
a2=0;
else if (v2>900)
a2=1;
if (v3<800)
a3=0;
else if (v3>850)
a3=1;
if (v4<820)
a4=0;
else if (v4>900)
16
a4=1;
if (v5<800)
a5=0;
else if (v5>900)
a5=1;
// 1 = Finger Bent
// 0 = Finger Straight
Serial.println("Playing....Stop(Change Needed)");
tmrpcm.play("Z.wav"); //the sound file "Z" will play each time the arduino powers up, or is
reset
lcd.clear();
// set the cursor to column 0, line 1
// (note: line 1 is the second row, since counting begins with 0):
lcd.setCursor(0, 1);
//Print a message to second line of LCD
lcd.print("Stop.(CN)");
delay(1000);
Serial.println("Done Playing");
}
else if (a1==1 && a2==0 && a3==0 && a4==0 && a5==0)
{
Serial.println("Playing....A");
tmrpcm.play("A.wav"); //the sound file "A" will play each time the arduino powers up,
or is reset
lcd.clear();
// set the cursor to column 0, line 1
// (note: line 1 is the second row, since counting begins with 0):
lcd.setCursor(0, 1);
17
//Print a message to second line of LCD
lcd.print("A.......");
delay(1000);
Serial.println("Done Playing");
}
18
TROUBLESHOOTING
1. As with any machine with moving parts, be careful not to let body parts get pinched.
While the hand has very little gaps and it may be difficult to get trapped in them, it
could potentially catch on clothing, rings, etc.
2. When operating this hand apply a maximum of 5 VDC to the LCD. Anything above
this voltage will destroy the LCD and will void the warranty.
3. When the hand is at rest and not performing all power should be turned off.
4. If excessive force is applied to any finger or thumb for any gesture, it may destroy the
sensors.
Many small test circuits had to be built on breadboards and tested in the project, under actual
circumstances, to make sure they would not fry.
CAUSE: Several components were not grounded and did in fact fry. A breadboard was also
melted. Also, it was difficult to manage a huge circuit on a bread board resulting in improper
connections of wires.
TROUBLE SHOOT: We managed working on it using jumper wires and reducing the use
of bread board wire.
19
2. FLEX SENSOR READING PROBLEMS
CAUSE: This problem maybe caused due to the wrong design of our analog circuit.
TROUBLE SHOOT :
CAUSE:
Display showing black boxes: This problem tends to occur if you have chosen a very
low value of the display contrast control resistor. Also, display may show black boxes
even when you are writing the data very fast onto the display.
Display showing wrong characters: The display shows wrong characters even when
you are sending the right ASCII values. This problem is because the display is not
receiving the proper data since there is a problem in the connection of data lines. There
might be a short circuit between adjacent data lines or it might also occur if some of
the data lines are not properly connected.
Display showing totally blank lines: This problem also occurs if you have set the
contrast control resistor value too high. So make sure you have selected the value to be
around 4.7K ohms which was default in our case.
Contrast OK, Delay OK but still no display: If you are not getting proper character
display even after setting the right contrast values and providing the right delay in code,
then there might be a chance that your controller's logic voltage levels are not
compatible with the display's.
20
TROUBLE SHOOT :
1. You are sending the proper commands. The address of each lines on LCD are not
consecutive and there are offsets. So we have to make sure if we are using the
appropriate LINE addresses as given in the LCD datasheet.
3. Hold the enable pin HIGH long enough for the display to latch the data in.
4. Wait for the LCD to initialise itself on power up and then start sending commands.
DB7 pin indicates LCD busy.
21
CONCLUSION
Sign language may be a helpful appliance to ease the communication between the deaf or mute
community and additionally the standard people. This project aims to lower the communication
gap between the mute community and additionally the standard world. The projected
methodology interprets language into speech. The system overcomes the necessary time
difficulties of dumb people and improves their manner. Compared with existing system the
projected arrangement is compact and is possible to carry to any places. This system converts
the language in associate passing voice that's well explicable by blind and ancient people. The
language interprets into some text kind displayed on the digital display screen, to facilitate the
deaf people likewise. In world applications, this system is helpful for deaf and dumb of us those
cannot communicate with ancient person. The foremost characteristic of this project is that the
gesture recognizer may be a standalone system, that's applied in commonplace of living. It’s in
addition useful for speech impaired and paralysed patient means those do not speak properly
and in addition used for Intelligent Home Applications and industrial applications.
22
FUTURE SCOPE
This project is cheap, efficient and portable. This system also uses simple techniques. Deaf and
dumb people are helped by this project doing communication in marking areas, public sectors,
working areas with others.
This project is also helpful and play major role in various fields such as Robotics, Biometrics,
Automation of industries, Musical instrument by replacing physical buttons and switches by
hand gestures.
To present the project 100% waterproof, some protected layers have to be placed in order to
save the circuit, battery and speaker from water.
To support more number of signs, and different languages mode the system can be extended.
In future, designing of a jacket which will be capable of detecting movement of animals can
also be done by improving the project.
System's efficiency can also be enhanced by doing different software development strategies
and various programming techniques.
23
REFERENCES
24