Vision onWay

Physical activity or exercise can improve your health and reduce the risk of developing several diseases like type 2 diabetes, cancer and cardiovascular disease, but it?s harder for blind and visually impaired people to walk and exercise freely in parks and jogging tracks on their own. There are sev

Project Title

Vision onWay

Project Area of Specialization

Artificial Intelligence

Project Summary

Physical activity or exercise can improve your health and reduce the risk of developing several diseases like type 2 diabetes, cancer and cardiovascular disease, but it’s harder for blind and visually impaired people to walk and exercise freely in parks and jogging tracks on their own. There are several barriers that often prevent people with impaired vision from engaging in physical activities. They include: Lack of motivation, energy, and fear of getting hurt, Lack of information to helpful resources, trouble learning new physical activities, that’s where our idea comes in Vision onWay. It is a mobile application that will help visually impaired people to keep them on track while exercising and walking. It will detect lanes using image processing and send a voice alert if they are getting off track or find any kind of obstacle in their following path. We will use Deep learning (CNN Algorithm) for lane and obstacle detection. Our Application will have audio-described workout videos. They will follow those videos and burn calories according to no of sets done.  It will contain fitness features like step counter, calories burned count, distance covered and time counter etc. Application will display current result after every walk and exercise, and monthly progress at the end of the month. Vision onWay will process voice commands given by user. Vision onWay ensures a safe and healthy lifestyle for visually impaired individuals. Our Application will be very simple and accessible for blind especially. Our motive is to promote cane free walk & exercise for blind people.

Project Objectives

Our objectives are to

•  Help visually impaired ones to keep themselves Healthy.
•  Make them able to track their fitness workout schedule.
•  Provide a safe environment during walk and workout.
•  Promote cane free walk and exercise for blind people.
•  Promote a sustainable environment for them.
•  Reduced inequalities.
• Promote Ease in Communication between app and blind individual.
• Motivate them regarding their physical fitness.
• Provide them ‘all in one health tool’.

Project Implementation Method

Implementation is divided into phases

Deep Learning Model:

In phase I we will develop Deep learning model. We will implement CNN Algorithm for lane and obstacle detection.First we will acquire dataset related to lane and obstacle detection from difference resources. Then import relevant libraries and APIs e.g. Numpy, OpenCV, Keras etc. Then we will import relevant dataset that we have collected. Apply CNN Algorithm According to given dataset.

Application

App will develop using Flutter SDK and Dart programming language having login, signup, profile page etc. Profile page requires field like name, height, weight, age, and show calculated BMI. App contain 2 main features Walking and Workout. After selecting walking section, recording screen will appear and do lane detection and obstacle detection in real time environment by using deep learning model (CNN Algorithm), generate alert through voice if they are getting off track or find any kind of obstacle in their following path. Current record will display on GUI that include time, no. of steps, calories burned, and distance covered. Workout: This section of application contain audio described workout videos and gifs. After workout, time and calories burned will display. App will process voice commands given by user.

Integration of Application and DL Model

In this phase we will integrate DL model with our application using integration technique.

We will convert DL file into .tflite format to ease the bulk of the file.

Integration with hardware

-We will connect camera with application via Bluetooth connectivity. Video will record through camera and process through app.

-We will also use wireless headphones connecting through Bluetooth for easy communication between blind people and App.

Benefits of the Project

Benefits of the Project:                           

• Vision onWay will ensure the fitness of blind people.
• The App will promote the cane free walk for blind individual.
• Vision onWay will promote the defensible and safe environment for them during walk and exercise.
• Vision onWay will reduce disparities between blind and other people.
• Audio-described workout videos will guide them during workout.
• Wearable Bluetooth camera helps them to walk comfortably without holding phone.
• Voice commands will help blind people to communicate with Application easily.
• Wireless headphones will deliver comfortable and pleasurable voice communication between System and Blind people.

Technical Details of Final Deliverable

We will use CNN (convolutional neural networks) for image processing of Lanes Detection & Object Detection. CNN, also known as ConvNets, consists of multiple layers and is mainly used for image processing and object detection. We'll use flutter SDK & dart programming language to build mobile application that will support iOS & Android and then integrate the DL model with Application. Bluetooth Camera, connected with the application, will record real time video of a track and pass to the application then CNN will do image processing with the video and process lane and obstacle detection, then generate a voice alert. Bluetooth earphones will connect with an application that communicate voice between Application and user. Voice alert, audio described video and any type of instructions will be heard through Bluetooth earphones. When they will start walking, step counter is activated, count steps and stop after walk and display no calories burned, weight and time. Auto-Described workout videos have no of sets to perform to burn calories. After workout calories-burned will display with no of sets done.

Final Deliverable of the Project

Software System

Core Industry

Health

Other Industries

IT

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others

Sustainable Development Goals

Good Health and Well-Being for People, Reduced Inequality

Required Resources

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