Search for vehicles using computer vision
Like Object Search, computer-vision-based Vehicle Search describes the ability to track a specific vehicle across multiple non-concurrent video feeds and sources. In real-world use cases, Number Plate Recognition alone often isn’t sufficient, with plates partially or fully obscured. To re-identify a single vehicle appearing in separate video frames, multiple machine learning algorithms often need to be applied, which may include a combination of high-level and fine-grained identification models.
Higher level machine learning models include Vehicle Detection (recognizing the presence of a vehicle) and Vehicle Type Recognition (distinguishing between cars, trucks, buses and the like). More granular models include Vehicle Make Recognition and Vehicle Color Recognition. Common computer vision use case for vehicle search include:
Security
Surveilling the movements of known or unauthorized vehicles throughout a facility.
Law enforcement
Searching for vehicles involved in crimes as part of post-incident investigations.
Emergency services
Locating a vehicle to assist with rescue or missing persons scenarios.
*Disclaimer: Outcomes described and depicted are illustrative in nature. Learn more.