Traffic Vehicle Analytics Dashboard

An end-to-end pipeline combining YOLOv8 vehicle detection with an interactive analytics dashboard built with Streamlit and Plotly.

Live Dashboard

Try it here: https://traffic-analytics-dashboard-yoji.streamlit.app/

What this project does

• Runs YOLOv8 on traffic camera images to detect and classify vehicles
• Processes detection results into time-series traffic data
• Visualizes patterns in an interactive Streamlit dashboard with filters

Dashboard Features

• KPI metrics (total vehicles, avg per interval, peak hour)
• Hourly traffic volume bar chart
• Vehicle type distribution pie chart
• Traffic flow time series
• Weekday vs weekend comparison
• Filterable by day, hour range, and vehicle type
• Raw detection data table

Tech Stack

• Python, YOLOv8 (Ultralytics)
• Streamlit, Plotly, pandas
• Kaggle (T4 GPU for detection)
• Streamlit Cloud (deployment)