Deep learning Assited Smart Glasses For Object Detection at low light conditions
We are building smart glasses that will work in Day, low light conditions, and in night Vision ( Total Darkness ) for object detection, and its audio feedback is given to blind people. That will help him to know more about its Environment. The whole system is based on Artificial intelligence where a
2025-06-28 16:26:05 - Adil Khan
Deep learning Assited Smart Glasses For Object Detection at low light conditions
Project Area of Specialization Artificial IntelligenceProject SummaryWe are building smart glasses that will work in Day, low light conditions, and in night Vision ( Total Darkness ) for object detection, and its audio feedback is given to blind people. That will help him to know more about its Environment. The whole system is based on Artificial intelligence where are collecting our custom dataset on night Vision and taring the model with the latest architecture Yolo V5/VR so that its mAP is higher than previous build smart glasses by doing outperforming research work.
Project Objectives- The main purpose of this product is to provide virtual eyes to a blind person
- To provide a high mean average precision while detecting the object
- Provide to gain a low latency rate while identifying objects.
- To replace traditional canes with smart wearable glasses to get in-depth detail of the Environment to the user
- Distance computation from camera to object(s)
- Audio voice feedback of the result
So the main functions of the prototype of the glasses are
- To get a Real-time video stream from the camera Then give it to the processing unit.
- It will have three mode
- For daytime
- ( Low light conditions )
- For Night Vision
- Perform object detection on the video.
a.If it day time it will Perform direct object detection
b. If it’s low light then perform low light image enhancement with the start of the Art Ai algorithms that perform Image enhancement and object detection.
c . Night Vision Mode: if there is total darkness then the night vision mode will be activated. the objects will be detected on our custom dataset of night vision .
2. After the object detection, the three will be voice feedback come in headphones/speakers to tell the BVI Blind Person Which object has been detected that helps them to navigate in its environment.
Benefits of the Project1. Huge Impact: Nearly 2 billion people are facing Vision problems According to WHO ( World health organization ) Out of which 1 billion people in the World cant see totally at day or Night at All. around 40 million people in the world are blind, while another 250 million have some form of visual impairment.
2. low cost: Our Product will cost 599$, which is too low as compared to other glasses in the World which is up to $3,292.
3. Night Vision: pervious Built glasses do not have night vision features in them we are providing Night vision features in our glasses.
4. High mAP research work: We have achieved highest Mean average person then pervious built glasses
5. Virtual EYE: we are providing a virtual eye to a blind person.
Technical Details of Final Deliverable1. Deep learning model based on transfer learning which will be able to detect and recognize objects in normal as well as low light conditions.
2. Software System developed using previously mention DL Model with the ability to Guide visually impaired people through audio instructions.
3. Prototype of Glasses which will demonstrate by using the processing power of the laptop for demonstration and for the final product we use the processing power of Jetson nano ..
Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Medical , Manufacturing , Health , Telecommunication Core Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT), Augmented & Virtual Reality, 3D/4D Printing, Robotics, Wearables and ImplantablesSustainable Development Goals Good Health and Well-Being for People, Affordable and Clean Energy, Decent Work and Economic Growth, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 79500 | |||
| NVIDIA Jetson Nano Developer Kit | Equipment | 1 | 35000 | 35000 |
| V380 Night Vision camera for Data Collection | Equipment | 1 | 5000 | 5000 |
| sq 11/23 mini day time camera for glasses | Equipment | 1 | 5000 | 5000 |
| Arducam 1080P Night Vision USB Camera | Equipment | 1 | 8500 | 8500 |
| Samsung 25W Battery Pack 20,000mAh | Equipment | 1 | 7500 | 7500 |
| Glasses | Equipment | 1 | 2500 | 2500 |
| Charger and Data Cable | Equipment | 1 | 3500 | 3500 |
| Headphones for Glasses | Equipment | 1 | 2500 | 2500 |
| All Miscellaneous Cost | Miscellaneous | 1 | 10000 | 10000 |