Desarrollo de un sistema de reconocimiento de placas de vehículos mediante transformada de Wavelet

Abstract

This research focuses on the application of the Wavelet transform technique for the automated recognition of vehicle license plates through their plate. This study aims to improve the identification of vehicle license plates through image processing and pattern recognition. The system consists of several components, including image acquisition through a security camera, image processing using Python libraries, and storage of results in a database. The Wavelet Transform is applied to improve the image features, allowing the detection of plate regions. The approach achieved an accuracy of 65.3% in plate identification, with potential areas for improvement identified as nighttime detection and variation in vehicle colors. The research highlights the power of the Wavelet transform in image analysis and signal processing, although limitations in image quality, lighting conditions, and plate variations can affect accuracy. Additional work could involve integrating the Wavelet transform with additional image processing or machine learning techniques to improve results, especially in challenging scenarios such as detection during nighttime hours.