YOLO-Traffic

Traffic analysis tool using YOLOv8 to count and track vehicles in video

Status: Research
Version: No tagged release
Platform: Desktop app
Source: github.com/Sherin-SEF-AI/YOLO-Traffic
License: Not specified

01 · Problem

Helps traffic planners and researchers gather traffic data from video more easily.

02 · Architecture

How it is put together.

A traffic analysis tool built on YOLOv8 object detection that counts and tracks vehicles in video footage. It detects vehicle types, counts vehicles crossing user-drawn regions, estimates speeds and produces basic traffic-flow visualisations. The README describes it as a work in progress; the repo holds several standalone Tkinter script versions.

FIG. 01Modules and data flows, from the repository layout and documentation.

1
Basic tool
Single-file Tkinter app using yolov8n
yolo-traffic.py
2
Analysis app
Extended Tkinter app with charts
YOLO-Analysis.py
3
Advanced app
Later versions with ROI drawing and exports
yolo-new.py

03 · Capabilities

Specification.

Detection: YOLOv8 (Ultralytics) vehicle-type detection
Counting: Vehicles crossing user-drawn regions of interest
Speed: Vehicle speed estimation
Output: Traffic-flow charts (Matplotlib) and CSV/JSON export
Interface: Tkinter desktop GUI
Runs on: Standalone Tkinter GUI scripts (tk.Tk() in yolo-traffic.py, yolo-new.py)

Stack