An Advanced ICTVSS Model for Real-Time Vehicle Traffic Applications

Area of Science:

• Computer Vision
• Intelligent Transportation Systems
• Smart City Technology

Background:

• Rising traffic accidents in Taiwan linked to increased vehicle numbers.
• Challenges in traffic safety include poor video quality, inefficient surveillance, and low camera resolution.
• Need for advanced real-time technologies to improve traffic accident reduction and evidence collection.

Purpose of the Study:

• To develop an intelligent cloud-based transportation vehicle surveillance system (ICTVSS).
• To create an advanced algorithm for dynamic license plate identification.
• To enhance traffic safety and support smart city initiatives.

Main Methods:

• Combined improved differential and active contour algorithms for license plate identification.
• Developed a novel dynamic license identification algorithm.
• Utilized high-resolution digital cameras for improved image quality.

Main Results:

• The proposed algorithm demonstrated good performance in locating multi-license plates.
• Effectively removed image noise and processed varying light conditions in complex environments.
• Achieved reliable license plate identification, providing a basis for judgment and improvement.

Conclusions:

• The ICTVSS system and its dynamic license identification algorithm are effective for smart city applications.
• The technology addresses existing challenges in traffic surveillance and evidence provision.
• Contributes to improving traffic safety and management through advanced technology.