Browse » Construction » Machinery
Top Machinery Datasets
Open source machinery computer vision datasets, pre-trained models, and APIs.
This project is trying to create an efficient computer or machine vision model to detect different kinds of construction equipment in construction sites and we are starting with three classes which are excavators, trucks, and wheel loaders.
The raw images (v1) contains:
- 1,532 annotated examples of "excavators"
- 1,269 annotated examples of "dump truck"
- 1,080 annotated examples of "wheel loader"
Note: versions 2 and 3 (v2 and v3) contain the raw images resized at 416 by 416 (stretch to) and 640 by 640 (stretch to) without any augmentations.
Here are a few use cases for this project:
Construction Site Monitoring: The "Test" model can be used to analyze images or video feeds from construction sites to identify and track the presence of different equipment and personnel. This can help site managers ensure safety protocols are followed and optimize the utilization of equipment on site.
Industrial and Warehouse Automation: Utilizing the "Test" model, businesses can automate certain processes such as inventory management or safety inspections by identifying and tracking the presence of equipment like wheel loaders, trucks, along with human personnel, effectively streamlining operations.
Equipment Maintenance & Inspection: The "Test" model can assist maintenance teams by recognizing specific heavy machinery (e.g., excavators, wheel loaders) and assessing whether or not the equipment is working correctly based on the state of the bucket (empty or full).
Training and Simulation: The "Test" model could be integrated into virtual reality (VR) or augmented reality (AR) training applications for construction or heavy equipment operators, enhancing the realism of the environment by accurately identifying and simulating elements such as trucks, excavators, and wheel loaders.
Search and Rescue Operations: The "Test" model can be used in search and rescue applications to quickly analyze drone footage or images from disaster zones, identifying the presence of humans, heavy equipment, or vehicles to prioritize rescue efforts and allocate resources effectively.