Smart Pothole Reporter Using Artificial Intelligence and PictoBlox

Road potholes are a major cause of vehicle damage, traffic congestion, and road accidents. Detecting potholes manually is time-consuming and inefficient. The Smart Pothole Reporter is an AI-powered system developed using PictoBlox that automatically identifies potholes through a camera feed and alerts the user in real time. The system uses image classification to distinguish between normal roads and potholes. This project demonstrates how Artificial Intelligence can be applied to improve road safety and support smart city infrastructure.

Materials Required

Laptop/Desktop

Webcam (built-in or external)

PictoBlox Software

1. Open PictoBlox
2. Create a New Project
3. Select:
4. Stage Mode
5. Python not required for this version

Create Two Classes

1. Pothole
2. Normal Road

Collect Images

1. 50–100 pothole images
2. 50–100 normal road images

Train Model

1. Open AI Model Trainer.
2. Upload images.
3. Train the model.
4. Deploy the trained model.

Create:

1. Pothole_Count

Add:

1. AI Model Extension
2. Text to Speech

Start Camera

When Green Flag Clicked

Turn Video ON

Set Pothole_Count to 0

Speak "Smart Pothole Reporter Started"

Detection Logic

Forever

Recognize Image

If Predicted Class = Pothole

Change Pothole_Count by 1

Speak "Warning. Pothole Detected Ahead"

Wait 3 Seconds

Else

Say "Road Normal"

Flowchart

Start

↓

Turn Camera ON

↓

Capture Road Image

↓

AI Prediction

↓

Pothole?

↙ ↘

Yes No

↓ ↓

Alert Continue

↓

Increase Counter

↓

Repeat

The Smart Pothole Reporter successfully detected potholes using the trained AI model. The system was able to distinguish between pothole images and normal road images with good accuracy. When a pothole was detected, an audio warning was generated and the pothole counter increased automatically. The project demonstrated the effectiveness of AI-based road monitoring.