3 of 11

Problem Statement

According to the World Health Organization, there are 285 million people who are considered visually impaired. About 245 million are diagnosed with low vision and 40 million are blind. In today’s fast-paced, high traffic world it can be difficult for the blind, or visually impaired to navigate areas.

Traditional forms of aid include visually impaired walking canes, and service dogs. However, these strategies have their limitations, and service dogs can be very expensive.

4 of 11

Problem Solution

Our system is designed to assist the visually impaired by increasing their knowledge of their surroundings. We aim to increase the efficiency of the user's mobility through audio aid and vibrational warnings. It can be used as another layer of affordable aid with traditional assistance canes or service dogs.
Designed to be lightweight and wearable for everyday use, it utilizes a camera and sensors in order to make users aware of a person or object that may be in front of them.

Personnel and Budget

Part	Cost
Raspberry Pi 3 Model B	Available in Lab
VL53L1X Time-of-Flight Distance Sensor	$11.95 x 6 = $71.70
Vibrating Mini Motor Disc (15 pack)	$12.99
Raspberry Pi Camera Module v1	Available in Lab
USB Battery Pack for Raspberry Pi - 4000mAh - 5V @ 1A	$24.95
ON/OFF Switch	$2.95
Chest Strap Harness	$8.99
Auxiliary Headphones	N/A (Already Owned)
TOTAL	$121.58

Hardware Design

System Design

Software Abilities

Software implementation is done in Python
Raspberry Pi utilizes OpenCV face recognition to track faces within video stream.

The device implements triangle similarity in order to track the distance from our camera to a detected object (face).

Milton -

Open cv has a database of samples that are used to train itself. In order to detect faces it needs a template for frontal face

- Example:

8.5 x 11in piece of paper ( W = 11) D = 24 in.front of camera, take a photo.
When we measure width of the piece of paper the perceived width of the paper is P = 248 pixels.
My focal length F is then:
F = (248px x 24in) / 11in = 543.45
Move my camera both closer and farther away from the object/marker, I can apply the triangle similarity to determine the distance of the object to the camera:
D’ = (W x F) / P
Again, to make this more concrete, let’s say I move my camera 3 ft (or 36 inches) away from my marker and take a photo of the same piece of paper. Through automatic image processing I am able to determine that the perceived width of the piece of paper is now 170 pixels. Plugging this into the equation we now get:
D’ = (11in x 543.45) / 170 = 35in
Or roughly 36 inches, which is 3 feet.

Software Abilities Continued

Flite Speech Synthesis Program converts distance measured into audio output (via headphones)

Vibrating haptic feedback discs are activated via pulse width modulation from detections by the time of flight sensors.

Conclusions

Overall, our design was successful and proved our concept, however there are a few improvements that could be made in future designs.
Designing the system using the most affordable sensors led to inaccuracies in our data, and reliability issues. Broken components caused setbacks and more labor hours. A major improvement would be to invest in higher quality components (i.e. camera and sensors).
Confident that our system will enhance the safety and awareness of its users