SMART OBJECT DETECTOR FOR VISUALLY IMPAIRED Govardhan.S.D 1, Kumar.G 2, Mariyappan.S 3, Naveen Kumar.G 4, Nawin Asir.J 5 Assistant Professor 1, Student 2,3,4,5 Department of Electronics and Communication Engineering Coimbatore Institute of Engineering and Technology, Coimbatore. Abstract: The main aim of this research work is to design and develop a low cost machine to the visually impaired for detecting the objects.the project is developed by using the arduino uno and sensors.in this project we are using two sensors i.e ultrasonic sensor and motion sensor.the ultrasonic sensor are used to recognize the obstacles and the motion sensor are used to determine the living things.finally the proposed system can able to detect an obstacles and living things by the voice recognisation system INTRODUCTION: Over 285 million people in the world were Blind or Virtually Impaired. They are restricted in both indoor and outdoor. The traditional Blind assistance is not suitable for today s world. Blind people have a very difficult time identifying items far in front of them. A smart device work at Blind assistance to solve all problem face the Blind or visually impaired people. One report says visually impaired people s foremost need is road crossing and identifying obstacle. The smart device fulfills the two needs with the use of smart sensors. The blind assistance navigates the visually impaired with the information about the surrounding environment and moving objects. The smart device consists of two major units. 1. Wearable smart sensor unit 2. Smartphone Many researchers have put forth efforts to develop systems to assist blind people in existing method. In the researches visually impaired people s foremost need is crossing road and identifying obstacle. The indoor navigation better and more easily navigate their surroundings using a sensor module act like a flashlight. This device searching a thing in 3D environment. This device connected to the computer to find characteristics position, orientation & navigation are send to computer answer through text to speech (TTS) engine But this device is limited to moving objects and colour objects are not detected The efficient way to identify the obstacle in indoor is digital sign method invented by Tjan in the year of 2005. To place the digital sign in all obstacle in the indoor like house. You have a digital sign receiver. If you nearer the digital sign the digital sign and digital sign receiver connected and alert like beep sound or vibrate. It has only useable for indoor and It is only works when nearer to object The RFID method was invented in the year of 2012 this is only use for identifying medicine in the cabinet for the visually impaired and the RFID and Robotic combined system was created in the same year and it is very efficient in indoor navigation and finding object for visually impaired but in outdoor the data collecting time is delay due to large object sensing @IJRTER-2017, All Rights Reserved 192
There are many other device are create for the helping the visually impaired people but these device are not provide the user to the distance of the object and moving object details that the eye usually provide. SYSTEM STRUCTURE and HARDWARE CONNECTION: In our device mainly needed a two main thing one is an inexpensive Smartphone and one wearable embedded device. These two devices are connected due to Bluetooth module or wifi module. This embedded sensor includes two sensors. This device mainly invented for the blind or visually impaired people the efficient way to feel, see, observe their environment and feel free to walk in outdoor and crowd place. The various sensors attached to the embedded microcontroller. It attaches an ultrasonic sensor and PIR motion sensor simultaneously. The smartphone is a mobile computer that now has enough data processing, computing, and storage capability and is an excellent human interface to present visually impaired people with critical information about their surroundings. The main part of the wearable part is sensors. The ultrasonic sensor is detecting an object to near them using of the ultrasonic wave. The motion sensor has to detect moving object of their surroundings during the change the surrounding temperature. The mobile app has encrypted the data given by the embedded device to audible information to the user. In our device has more advandages. It detects both obstacle and living thing like human being. It works when the object is 3m distance.it is efficient in both indoor and outdoor. The Arduino UNO which allow to quick and easy response to both sensor data and this UNO type arduino board has very low cast like RS 500. Program in Arduino is simple embedded C language and the coding is very easy. Arduino UNO has 6 analog I/O pin, 6 digital I/O pin and 6 analog pwm pin available and 2 supply pin (5V, 3.5V). It has working in a 9V power. The ultrasonic sensor has a one transmitter called trig pin and a receiver called an echo pin. It has detected an obstacle in the range of 4 meter. A calculation calculates the distance of the obstacle it provides in the program the arduino UNO board. Distance=duration*0.034/2 It has one transmitter and receiver. Transmitter transmits the short burst of sound in particular direction. It receives with the time interval of t. In the program we have specified a 3 distance like 2 meter, 1.5 meter and 1 meter. If any obstacle enter the region of 2 meter it alert the visually impaired People and it increase the alert in 1.5 meter and high alert in 1 meter. It operates at a frequency of 40 KHZ. It operates by generating high frequency pulse @IJRTER-2017, All Rights Reserved 193
of sound wearable device these two unit talk each other via Bluetooth. Different software available in Smartphone and wearable device to communicate each other. Eclipse Integrated Development Environment is the android development tool utilizes all the android functionality. Motion sensor are used to detect moving object particularly people. It can detect up to a distance of 7 meters. It is wide use in domestic and commercial application. The PIR motion sensor is a passive infrared sensor it only detect a temperature change in surrounding at is use to the motion detection of infrared beam emitted from the living things. The Bluetooth module has use in HC-05 type Typical -80dBm sensitivity Up to +4dBm RF transmit power Low Power 1.8V Operation,1.8 to 3.6V I/O With integrated antenna. Cost of the Bluetooth module is Rs 250 Two major units are the embedded wearable sensor and a low cost Smart phone. The embedded wearable small sensor has four major units Power has supply with 9V mini size battery and the Microcontroller of Arduino UNO and some sensors has the major role of the collecting information of surrounding environment and finally a Bluetooth module or wifi module used to communicate the wearable micro controller and a smart phone. A smart phone receives data obtained from the wearable smart sensor unit processes the data in real time.sensor detect on object in front of a user with in 3m. The smart phone send beep to the user at high frequency as an object get closer. Smart phone and SMART APP: The application begins with a click on the application icon to launch it. Once it is launched, the user is provided with a variety of Bluetooth devices that can be programmed to remember previously used devices as well as automatically link with the sensor Once linked the mobile application continuously receive the data for sensors. The sensors sense the surrounding of 1 to 10 feet. This method is very efficient to visually impaired people. If the obstacle come near the visually impaired people the sensor identify and transfer to the smart mobile app, The app alert the visually impaired people with the beep sound or the voice command. TESTING: The application was tested with the addition of the audio prompt for the object distance in real time scenario. Beep frequency range plotted in above table. The following was used as the benchmark: a beep of 300 Hz (frequency) for objects beyond 90 @IJRTER-2017, All Rights Reserved 194
in, 400 Hz for objects between 60 and 90 in, 500 Hz for objects between 30 and 60 in, and 600 Hz for objects closer than 30 in. With the optimizing factor applied. In indoor navigation the perfect beep for every obstacle it s very useful for both indoor and outdoor. CONCLUSION: A smart assistant system for the visually impaired was developed, implemented, and tested. This system is a combination of a smart object-detecting wearable module and a smartphone, which provides critical environmental information for the visually impaired to better navigate his surroundings. Our system differs from existing work in that we took the advantage of the powerful mobile smartphone and well-developed smart sensors. Future research primarily focus on developing the application system outside of a controlled laboratory environment Designing the system to be particularly robust in every day user interaction real word along with real time. The current application combined with future work will allow the visually impaired a gain a novel sense of the world around them. Read more about it World Health Organization (WHO). (2014). Visual impairment and blindness [Online]. Available: http:// www.who.int/mediacentre/factsheets/ fs282/en/ S. Alwi and N. Mohamad, Survey on outdoor navigation system needs for blind people, in IEEE Proc. Student Conf. Research and Development, 2013, pp. 144 148. B. S. Tjan, P. J. Beckmann, R. Roy, N. Giudice, and G. E. Legge, Digital sign system for indoor wayfinding for the visually impaired, in IEEE Proc. Computer Society Conf. Computer Vision and Pattern Recognition, 2005, pp. 30 37. V. Kulyukin, C. Gharpure, J. Nicholson, and S. Pavithran, RFID in robot-assisted indoor navigation for the visually impaired, in IEEE Proc. Intelligent Robots and Systems, 2004, pp. 1979 1984. R. Velazquez, Wearable assistive devices for the blind, in Wearable and Autonomous Biomedical Devices and Systems for Smart Environment. Berlin, Germany: Springer, 2010, pp. 331 349. J. Rodriguez, J. Yebes, P. Alcantarilla, L. Mergasa, J. Almazan, and A. Cela, Assisting the visually impaired: Obstacle detection and warning system by acoustic feedback, Sensors, vol. 12, no. 12, pp. 17476 17496, Dec. 2012. J. D. Hub and T. Ertl, Design and development of an indoor navigation and object identification system for the blind, in Proc. ACM SIGACCESS on Accessibility and Computing, 2004, pp. 147 152. About the authors David Zhou (davidzhouabc@gmail. com) is a senior at Cypress Ranch High School and a computer science researcher at the Department of Computer Science at Prairie View A&M University. His research interests are smartphone applications and computer vision. Yonggao Yang (yoyang@pvamu. edu) earned his Ph.D. degree in information technology from George Mason University. He is a professor and department head in the Department of Computer Science at Prairie View A&M University. @IJRTER-2017, All Rights Reserved 195