click here to download project abstract
click here to download project base paper
–
ABSTRACT:
Although artificial vision systems could potentially provide very useful input to
assistive devices for blind people, such devices are rarely used outside of laboratory
experiments. Many current systems attempt to reproduce the visual image via an
alternative sensory modality (often auditory or somatosensory), but this dominant
“scoreboard” approach, is often difficult to interpret for the user. Here, I propose to
offload the recognition problem onto a separate image processing system that then
provides the user with just the essential information about the People in their
surroundings using Facial Recognition . Specifically, I am going to work on image
processing algorithm not only robustly, precisely, but something which rapidly
recognize the people around them. In addition, of Facial Recognition, I am also
using Google- Text-To Speech/PYTTS which basically helps my Algorithm to
provide a user friendly experience.
This is a helpful aid for visually impaired people and people with disabilities. Facial
Recoginition will generate an output string and converts them into audio speech
using pre and post processing with gTTS (Google Text to Speech). Google is used
as its platform for virtual assistant which can be used in day to day life activities like
checking mails, weather-forecast, news etc., further using Google Assistant, and
python language we implement a voice based home automation. The major
objective of this project is to help visually impaired by using various fields of
technology.
This is a useful guide for outwardly debilitated endlessly individuals with
incapacities. Facial Recognition will produce a result string and converts them into
sound discourse utilizing pre and post handling with gTTS (GoogleText to Speech).
Google is utilized as its foundation for remote helper which can be utilized in
everyday life exercises like checking sends, climate estimate, news and soon, further
utilizing Google Assistant, and python language we carry out a voice based
home mechanization. The significant target of this task is to help outwardly
hindered byutilizing different areas of innovation.
INTRODUCTION:
As per the UNESCO and the World Health Organization (WHO), around 300
million individuals all over the planet are assessed to be outwardly
debilitated, out of which 87% live in non-industrial nations. Due to outwardly
impairedness individuals can’t perceive individuals which makes the
significant issue for outwardly disabled populace which are dealing with
numerous issues, A significant issue for an outwardly debilitated or outwardly
hindered individual (BVI) to associate with the world to share information .
India has the largest share of the world’s blind? The estimated size of the
severely vision impaired Indians is 1 crore ( 10 million ) persons. Added to
this are another 1 crore persons who are Low Vision ( moderately vision
impaired who too are disabled to work and live like normal sighted
people).Visually Impaired people face a lot of difficulties.
There have been developments on mobile phones and computers that assist a
blind person by combining computer vision tools with other existing
expedient products such as Optical Character Recognition (OCR) system.
This can be diminished by executing a mix of flexible innovation for example
Facial Recognition and Text to Speech Engine which is a great device for
individuals carrying on with their everyday life.
This Prototypes works in two phase i.e pre- handling and post handling where
extraction, division and acknowledgment happens where the last data is
gotten which is then changed over into discourse or sound result by utilizing
different discourse motor, for example, gTTS, E-Speak however gTTs is
more exact than TTs. The preffered system assists blind people by capturing
the text and then by reading it to them. Extracting the text present is enacted
with OCR. It is a tactic for transformation of images of writings on a label,
printed books etc.OCR replaces binary images with texts and also detects
white spaces. It also parses the integrity of the recognized text.
Abstract:
The “Vision for Visually Impaired” project aims to enhance the daily lives of visually impaired individuals by leveraging Python-based computer vision techniques. The existing system relies on traditional aids like canes and guide dogs, offering limited assistance. The proposed system utilizes advanced computer vision algorithms to provide real-time environmental awareness, object recognition, and navigation assistance.
Existing System:
The current systems for visually impaired individuals are predominantly hardware-based, lacking in intelligent real-time feedback. Traditional aids, such as canes and guide dogs, have limitations in providing comprehensive environmental information.
Proposed System:
Our proposed system employs computer vision algorithms to analyze live camera feed, identify objects and obstacles, and provide auditory or haptic feedback to the user. The system aims to enhance independence and safety by offering a more intelligent and adaptive solution for navigation.
System Requirements:
- Python 3.x
- OpenCV for computer vision
- TensorFlow for deep learning
- Text-to-speech libraries for auditory feedback
- Compatible camera or smartphone with a camera
Algorithms:
- Object detection using Convolutional Neural Networks (CNN)
- Image processing for feature extraction
- Machine learning for personalized navigation
Hardware Requirements:
- Camera (integrated or external)
- Smartphone or Raspberry Pi for processing
Software Requirements:
- Operating System: Linux or Windows
- Python IDE (e.g., Jupyter Notebook)
- OpenCV, TensorFlow, and other relevant libraries
Architecture:
The system follows a modular architecture comprising data acquisition, image processing, object detection, and user feedback modules. It integrates seamlessly with the user’s device, providing real-time assistance.
Technologies Used:
- Python
- OpenCV
- TensorFlow
- Text-to-speech libraries (e.g., pyttsx3)
Web User Interface:
The project offers a user-friendly web interface for configuring preferences, reviewing past navigations, and updating the system. The interface ensures accessibility for both visually impaired and sighted users.
Class Diagram
Sequence Diagram
Use Case Diagram
Activity Diagram
Collaboration Diagram
Deployment Diagram
