Object detection and warning system for visually impaired people

Visually impaired people face various troubles in their day by day life, and innovative mediations may help them to address these difficulties. As per measurements of the World Health Organization (WHO), universally, in any event 2.2 billion individuals have a dream debilitation or visual impairment

Project Title

Object detection and warning system for visually impaired people

Project Area of Specialization

Artificial Intelligence

Project Summary

Visually impaired people face various troubles in their day by day life, and innovative mediations may help them to address these difficulties. As per measurements of the World Health Organization (WHO), universally, in any event 2.2 billion individuals have a dream debilitation or visual impairment, of whom at any rate 1 billion are visually impaired. The proposed framework assists the outwardly debilitated individuals in moving and make them autonomous of any outside help. This system proposes a man-made reasoning based completely programmed assistive innovation to perceive various items, and hear-able sources of info are given to the client continuously, which gives better comprehension to the visually impaired people about their environmental factors. A profound learning model is prepared with different pictures of items that are exceptionally applicable to the outwardly weakened individual. Notwithstanding computer vision-based procedures for object acknowledgment, a distance-estimating sensor is coordinated to make the gadget more extensive by perceiving snags while exploring starting with one spot then onto the next. The hear-able data that is passed on to the client after scene division and hindrance recognizable proof is upgraded to get more data in less an ideal opportunity for quicker handling of video frames. The normal exactness of this proposed technique is 95.19% and 99.69% for object discovery and acknowledgment, individually. The time multifaceted nature is low, permitting a client to see the encompassing scene continuously.

Project Objectives

Vision impairment is one of the major health challenges in the world. Perception is one of the essential human senses, and it plays a considerable role in human perception about the surrounding environment.According to World Health Organization (WHO), nearly 1.3 billion people live with some forms of vision impairment while 36 million people are completely blind (Blindness and vision impairment). Although many suggestions and proposals to reduce the causes of such sensation (Kartubi et al., 2020), the number of VI persons in the world is still escalating. This wide-spread dilemma pushes researchers to look for solutions or procedures to help the VI persons. The recommended solutions emphasis on how to make VI able to steer and recognize objects without the support of others. Nevertheless, most of the proposed techniques suffer either from technical limitations or practicality limitations (such as non-portability, size, high power consumption and limited support).

Our proposed system helps VI people in their day by day life by allowing them to acquire adequate data about items situated in a climate. The proposed framework shouldn't be connected on the web and it can get a high pace of effective article acknowledgment as differentiated to other existing on the web strategies. The proposed framework is viable, versatile simple to-utilize and devours substantially less force contrasted with other existing methods.

• It works as a navigation device for blind people.
• The system can be used to navigate by everyone not only visually impaired under certain circumstances, like foggy morning with low visibility.
• The system can also be useful for patients suffering with various eyes ailments
• Guide blind people.
• Alert through voice based system
• Obstacle detection
• Low power consumption

Project Implementation Method

The proposed system is set up in such a manner where an android application will capture real-time images and will send it to a Laptop based Networked Server where all the calculations happen. The Laptop Based Server will utilize a pre-prepared SSD identification model prepared on COCO DATASETS. The COCO dataset is labeled, providing data to train supervised computer vision models that are able to identify the common objects in the dataset.It will at that point test and the output class will get identified with a precision measurements. After testing with the assistance of voice modules the class of the object will be changed over into a default voice notes which will at that point be sent to the visually impaired persons for their help. Alongside the object identification, we have utilized analertsystem where estimated distance of the object from the VI person will be determined. To calculate the distance we used a sensor.If the Blind Person is exceptionally near the object or is far away at a more secure spot, it will produce voice-based output along with distance units.

Benefits of the Project

An assistive framework is suggested for visually impaired  people through which they can recognize their environmental factors and articles progressively and travel through uninhibitedly. Profound learning-based article openness, in help with different distance sensors, is utilized to make the client perceptive of snags, to give safe route where all data is given to the client as sound. Our high level framework utilizes a camera to hold onto ongoing pictures before the clients. The proposed framework shows a 5% blunder edge in distinguishing the distance and length of recognized items. The general correlation can profit numerous different explores that need to utilize object recognition strategies. The proposed framework has numerous preferences over the other two existing frameworks, for example, disconnected use, conveyability, simple to-utilize interface and minimal effort.

Technical Details of Final Deliverable

• Guide blind people.
• Alert through voice based system
• Obstacle detection
• Low power consumption
• It works as a navigation device for blind people.
• The system can be used to navigate by everyone not only visually impaired under certain circumstances, like foggy morning with low visibility.
• The system can also be useful for patients suffering with various eyes ailments

Final Deliverable of the Project

Hardware System

Core Industry

Health

Other Industries

Medical

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT)

Sustainable Development Goals

No Poverty, Good Health and Well-Being for People, Industry, Innovation and Infrastructure, Life on Land, Partnerships to achieve the Goal

Required Resources

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