ENVISION EYE

One of the  Innovative project named as Envision Eye is being made for visually impaired for blind people as we know that issue of visual impairment or blindness is faced worldwide. If we go for Regional Differences, the prevalence of visual impairment in low & middle income region is four

Project Title

ENVISION EYE

Project Area of Specialization

Artificial Intelligence

Project Summary

One of the  Innovative project named as Envision Eye is being made for visually impaired for blind people as we know that issue of visual impairment or blindness is faced worldwide. If we go for Regional Differences, the prevalence of visual impairment in low & middle income region is four time higher than in high income region. Blind  people  generally rely on stick, guide dogs, screen-reading software,  and glasses to assist  them for mobility, in order  to help the blind people the visual world has to be transformed into the audio world with the potential to inform  them  about  objects  as  well  as  their  spatial  locations. Therefore,  we  propose  to  aid  the  visually  impaired  by introducing  a system  that  is most  feasible, compact, and cost-effective. So, we implied a system that makes use  of  Raspberry Pi  in which  you  only look  once (TensorFlow) machine  learning algorithm  trained  on  the  coco  database  is  applied.  The experimental  result  shows  TensorFlow achieves  state-of-the-art results of 85% to  95% on  overall performance,  100% (person, chair, clock, and cell-phone etc )  recognition accuracy.  This system not only provides mobility  to the  visually impaired  with that it provides  the term that  ahead is  an XYZ  object rather  than a sense of obstacle which makes possible for blind person person to become independent. According to our proposed project it captures real-time images, then images are  pre-processed according to their  background  and  foreground they are separated and then the DNN (Deep Neural Network module ) with a pre-trained Tensor Flow model is applied resulting in feature extraction. The extracted features are matched with known object features to identify the objects. Once the object is successfully recognized, the object name is stated on bounding boxes as a text then the text is converted in to voice output with the help of text-to-speech conversion. The key contributions of the project includes,

• Efficient object detection and recognition of object/obstacles for visually impaired people to independently access familiar and unfamiliar environments and avoid dangers.
• Offline text-to-speech conversion and speech output.
• Weight approximately 150-200 gram so feasible for blind person to hang on Neck.
• Long Lasting Battery Timings.

Project Objectives

The Objectives of this project is to help visually impaired people, Basically the  concept behind this Eye is to assist the Blind person. The project prototype includes a custom designed border platform for Obstacle Detection, an Eye which is connected with a camera and earpiece, connected to the single board computer the Raspberry Pi 4B,  Python software  for raspberry pi is used to implement the modelling and coding of the Object Detection .

The objectives of the project:

• To design and implement smart Eye that could be used by the visually impaired & blind people easily, with an emphasis on cost-effectiveness.
• To prove the feasibility of the image detection techniques with the audio description as a tool helping the visually impaired people and giving them more independence in their daily  life.
• Help the visually impaired people easily Recognize between  the person and other daily routine objects.
• The main objective of this Eye is to identify and locate one or more effective targets from still image or video data.
• The main objective of this Eye is it weight around 150 grams which is main motive of my project to make it feasible for blind person to hang it on neck easily.
• This Envision Eye is having a good battery back-up.

Project Implementation Method

The microcontroller receives the images from the surrounding, processes them, and returns the result in audio format.  Smart Eye hardware has a Speaker/Earphone for direct output in the form of audio connection to convey audio results  to users.

The working of our project  is as follows: first, the user makes a connection between a smart Eye through Raspberry pi and earphone. Following this, the system can send a request to allow smart eye to capture images, and microcontroller receives the images. In this scenario, the power consumption of smart Eye can be reduced, which is much more efficient than continuous video scanning. Thereafter, the results from the artificial intelligence server are delivered in voice feedback via earphones or speaker . Further, Raspberry pi camera capture the image and send it to the controller which process the image and send the voice feedback hence in this way we will  perform object detection and recognition . In addition, this increases the battery life of smart Eye because they are used only for capturing images.

Benefits of the Project

The benefits of Envision Eye is to assist blind person and make him independent for his daily life routine. This project is intended to help these type of people to widen their scope of independence by giving them a description of the live scenes delivered in an audio format using an earpiece. The project is implemented using Python as a main software program, and the single board computer the raspberry pi  as a platform. The project is also using the raspberry pi camera to capture real time Picture and an earpiece to voice out the descriptions. It has wide scope and it has various application like automated driving, object detection, healthcare and many more. For example, by using deep learning detection scope will be presented as outcome of the project .

• Its is AI based Eye and detects objects/ obstacles that comes in blind person way and send a feedback to blind person through earphone in voice.
• The  visually impaired people will be able to recognize a object in front of them through object detection technique.
• Deep Learning is a subfield of Machine Learning. It has wide scope and it has various application like automated driving, object detection, healthcare and many more. For example, by using deep learning.
• Envision Eye , basically known as Artificial Eye use the power of speech to make life more accessible for blind person.
• Envision Eye is Product of my innovative research project that harnesses the power of Artificial intelligence to open up the visual world by describing nearby objects and person etc.
• Long lasting battery make it feasible for blind person.

Technical Details of Final Deliverable

The technical details of our FYP includes Components which are;

• Raspberry_pi 4B single board computer: the heart of the processing , executing the program and integrating the inputs and outputs of the system.
• Raspberry pi camera: for image acquisition, it is a 5 megapixel camera with resolution 2592 * 1944 pixels connected to the raspberry pi via ribbon cable.

• Tony Python for programming.
• Earpiece: To deliver the audio output to the blind Person.

The project is implemented using Python as a main software Program(tool), the single board computer raspberry pi 4B as platform, raspberry pi camera to capture real time picture. The implementation is done using OpenCV and TensorFlow libraries. Furthermore I have used Deep Neural Network  algorithm for obstacle detection. The picture of  my FYP Envision Eye is as below.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Medical

Other Industries

Health

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others

Sustainable Development Goals

Good Health and Well-Being for People, Gender Equality, Affordable and Clean Energy, Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com