Technologies Involved:
PYTHON
Area Of Work: Computer Vision
Project Description

A leading industrial automation company specializing in warehouse operations approached Oodles for a smart alternative to manual ceiling height measurements. Their need was clear: a fast, camera-based tool that eliminates the use of laser instruments. A YOLOv8-powered image analysis solution was delivered to detect and measure ceiling heights with accuracy and ease.

Scope Of Work

The client needed a computer vision system to detect ceilings, floors, and propellers from images and compute real-time height data. Areas of work included image data annotation, YOLOv8 model training, algorithm design for pixel-to-metric conversions, and support for both calibration-assisted and calibration-free workflows.

Our Solution

To solve the client’s challenge, a custom AI-powered measurement tool was developed using YOLOv8 and Python. 

An algorithm was integrated to convert detected object positions into real-world height data. This supported two workflows:

  1. With Calibration – Uses the Coca-Cola bottle for accurate pixel-to-cm scaling.
     
  2. Without Calibration – Estimates height using relative object proportions and bounding box logic.

Key Features Implemented:

  • YOLOv8-based object detection for ceiling, floor, and propellers
  • Roboflow-assisted image annotation pipeline
  • Pixel-to-metric height estimation algorithm
  • Dual-mode operation: calibrated and non-calibrated. 

Related Projects