基於多重亮度紅外線照明之夜間車牌及人臉辨識系統

Abstract

夜間監控變得越來越普及，而車牌辨識和人臉辨識是夜間監控的重要功能。然而夜間影像常常會因為光源不足、過曝以及逆光等情況變得不清楚，為了解決這些問題，我們使用多重亮度紅外線打光器來補助夜視攝影提供不同亮度的影像。由於多重亮度紅外線攝影會產生不同亮度的影像導致有不同結果，所以分析這些結果並整合為一即為本論文研究之目的。實驗結果顯示，多重亮度照明確實有助於提升夜間車牌辨識及人臉辨識的辨識率。

Nighttime video surveillance has become more and more popular. License plate recognition and face recognition are two important applications in nighttime video surveillance. However, nighttime videos are unclear due to insufficient illumination at far distance, overexposure at near distance and backlight situation. To solve these problems, we adopt a new multi-intensity infrared illuminator as a supportive light source to provide multiple illumination levels. Because there are imaging results with multiple intensities, analyzing and integrating these results is the purpose of this thesis. Experiment results show that multi-intensity infrared illumination can indeed improve the performance of license plate recognition and face recognition.