Indoor Object Detection and Recognition for Visually Impaired Users.ara
According to World Health Organization (WHO) data, there are about 10% population of the world that has eye related diseases. Majority of them are having color blind problems, which means that they are visually impaired. Organizing and Navigating things for a visually impaired users is such a challe
2025-06-28 16:33:06 - Adil Khan
Indoor Object Detection and Recognition for Visually Impaired Users.ara
Project Area of Specialization Artificial IntelligenceProject SummaryAccording to World Health Organization (WHO) data, there are about 10% population of the world that has eye related diseases. Majority of them are having color blind problems, which means that they are visually impaired. Organizing and Navigating things for a visually impaired users is such a challenging job to do. Even the more difficult job for a visually impaired user is to distinguish the different items around him/her.
The aim of the proposed project (Indoor object detection and recognition for visually impaired Users) is to help a visually impaired person in order to organize, navigate, and distinguish things around him/her. The mobile phone camera will predict the object and the proximity of the objects. The object detection module will help a visually impaired user to navigate and/or organize things/objects around their surroundings. This approach will be achieved by using an android smartphone. The visually blind user will only have to start noticing the surroundings through the mobile’s camera and the camera will detect and recognize the objects and will convert the detected objectss into voice as a result.
The main objective of the project is to help a visually impaired user by providing him/her a digital solution to recognize the objects in their surroundings.
Besides that, the objective of this project is to build an app which helps the blinds in following ways.
• To detect the object
• To recognize the objects
• To convert detected object into speech
• To recognize text into speech
The proposed project (Indoor object detection and recognition for visually impaired Users) will be implemented by using AI (Machine learning and Deep Learning) aka. Neural Networks. Various machine learning methods and frameworks are available to train/process the objects. For example, YOLO V. 3 or TensorFlow are one of the best machine frameworks to process/train the objects. TensorFlow or Yolo v3 could be used as a backend machine learning mechanism.
Thresholding technique for background and foreground separation, and by extracting the essential objects from the image by YOLO V. 3 or TensorFlow machine learning frameworks. And the matching extracted objects will be compared to the database of objects, e.g. pen, book, mobile etc. Android Studio will be used as frontend for the development of Android Application.
Benefits of the ProjectOne obvious and an immediate benefit of the project is that object detection and recognition method will be more robust in viewpoint. Applying Artificial Intelligence and Machine Learning techniques will enhance our dexterity in this very area of study.
Secondly, the project idea is inspired by helping the visually impaired person by providing them a digital solution in order to detect and recognize item in their surroundings. At the end we will be able to provide them a digital solution to their problem.
The final deliverables are listed as follows:
An Android Application will aid blinds and visually impaired persons in detection of objects in the surroundings using android smartphone camera.
A final year project thesis / report to fulfill the requirements of the Bachelors of Computer Science at University of Sindh, Mirpurkhas
Final Deliverable of the Project Software SystemCore Industry HealthOther Industries Security Core Technology Artificial Intelligence(AI)Other Technologies OthersSustainable Development Goals Good Health and Well-Being for People, Decent Work and Economic GrowthRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 40100 | |||
| Samsung Galaxy J7 Pro | Equipment | 1 | 35000 | 35000 |
| Printing cost | Miscellaneous | 3 | 700 | 2100 |
| Thesis binding cost | Miscellaneous | 3 | 1000 | 3000 |