Current Issue
Romanian Journal of Information Technology and Automatic Control / Vol. 35, No. 2, 2025
Improving low-light image quality for object detection and license plate recognition
Rishab SRIDHAR, Rohith Arumugam SURESH, Shreejith BABU, Priyadharsini RAVISANKAR
Enhancing low-light image quality is crucial for object detection and license plate recognition in surveillance and security applications. Poor illumination degrades image clarity, making accurate recognition difficult. This paper investigates a combination of image enhancement techniques - including Unsharp Masking, Gamma Correction, Gaussian Blur, and Histogram Equalization - to improve visibility and recognition accuracy in low-light conditions. The performance of these methods is quantitatively evaluated using Laplacian variance as a measure of image sharpness and clarity. Experimental results demonstrate that Gamma Correction applied to Unsharp Masking and Histogram Equalization significantly enhances image quality, enabling the accurate extraction and recognition of license plate numbers. The proposed approach successfully extracts and recognizes license plate numbers under poor lighting conditions, demonstrating its effectiveness for real-world surveillance applications.
Keywords:
Histogram Equalization, License Plate Recognition, Low-Light Image Enhancement, Object Detection, Surveillance Systems, Unsharp Masking.
CITE THIS PAPER AS:
Rishab SRIDHAR,
Rohith Arumugam SURESH,
Shreejith BABU,
Priyadharsini RAVISANKAR,
"Improving low-light image quality for object detection and license plate recognition",
Romanian Journal of Information Technology and Automatic Control,
ISSN 1220-1758,
vol. 35(2),
pp. 87-98,
2025.
https://doi.org/10.33436/v35i2y202507