OBJECT DETECTION FOR LOW-LIGHT ENVIRONMENT USING MULTISCALE RETINEX

Anthonius Adi Nugroho

doi:10.34288/jri.v8i2.493

Authors

Anthonius Adi Nugroho Universitas Kristen Duta Wacana

(*) Corresponding Author

DOI:

https://doi.org/10.34288/jri.v8i2.493

Keywords:

Object Detection, Low-Light Enhancement, Multiscale Retinex, SSD MobileNetV2

Abstract

Object detection is a critical task in computer vision, yet its performance degrades significantly under low-light conditions due to loss of detail and diminished features. This study proposes an image enhancement framework to improve detection robustness in challenging lighting. The methodology integrates Multiscale Retinex (MSR) for image enhancement and SSD MobileNet V2 for object detection. MSR was configured with optimal parameters (scale1:10, scale2:60, scale3:180, σ:100, β:30) to enhance brightness while preserving crucial image details. The experimental results demonstrate that Retinex correction is highly effective in extreme low-light scenarios. In 0 lux conditions, where objects were completely undetectable without processing, the proposed method enabled detection with confidence levels between 62% and 96%, yielding an average accuracy increase of 50%. In 15 lux conditions, accuracy improved by 6.6%. However, the system degraded at intensities above 25 lux, suggesting that the enhancement is most beneficial in near-dark environments. In conclusion, Multiscale Retinex significantly enhances the capability of SSD MobileNet V2 for object detection in environments with illumination below 77 lux. This approach provides a viable solution for improving the reliability of surveillance and automated systems operating in unpredictable lighting.

Downloads

Download data is not yet available.

References

Al-refai, G., ElMoaqet, H., Al-Refai, A., Alzu’bi, A., Al-Hadhrami, T., & Alkhateeb, A. (2025). Two-Stage Object Detection in Low-Light Environments Using Deep Learning Image Enhancement. Peerj Computer Science, 11, e2799. https://doi.org/10.7717/peerj-cs.2799

Bel, A., Sbert, C., & Morel, J. (2014). Multiscale Retinex. Image Processing On Line, 4, 71–88.

Estrada, J., Paheding, S., Yang, X., & Niyaz, Q. (2022). Deep-Learning-Incorporated Augmented Reality Application for Engineering Lab Training. Applied Sciences, 12(10), 5159. https://doi.org/10.3390/app12105159

Gasparyan, H., Hovhannisyan, S., Babayan, S. V, & Agaian, S. С. (2023). Iterative Retinex-Based Decomposition Framework for Low Light Visibility Restoration. Ieee Access, 11, 40298–40313. https://doi.org/10.1109/access.2023.3269719

Hanumantharaju, M. C., Ravishankar, M., Rameshbabu, D. R., & Ramachandran, S. (2011). Color Image Enhancement using Multiscale Retinex with Modified Color Restoration Technique. Emerging Applications of Information Technology.

Howard, A., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. https://doi.org/10.48550/arxiv.1704.04861

Huang, W., Zhu, Y., & Huang, R. (2020). Low Light Image Enhancement Network With Attention Mechanism and Retinex Model. Ieee Access, 8, 74306–74314. https://doi.org/10.1109/access.2020.2988767

Kee, E., Chong, J. J., Choong, Z. J., & Lau, M. (2024). Object Detection With Hyperparameter and Image Enhancement Optimisation for a Smart and Lean Pick-and-Place Solution. Signals, 5(1), 87–104. https://doi.org/10.3390/signals5010005

Khammar, M. M. (2024). Visual Intelligence: Machine Learning Approaches to Image Filtering and Identification. International Journal of Scientific Research in Engineering and Management, 08(04), 1–5. https://doi.org/10.55041/ijsrem29940

Kim, Y.-J., Son, D.-M., & Lee, S.-H. (2024). Retinex Jointed Multiscale CLAHE Model for HDR Image Tone Compression. Mathematics, 12(10), 1541. https://doi.org/10.3390/math12101541

Lan, Z., & Guo, Y. (2023). Low Illumination Image Enhancement Based on Retinex Theory. 35. https://doi.org/10.1117/12.3005820

Land, E. H., & McCann, J. J. (1971). Lightness and Retinex Theory. Journal of the Optical Society of America, 61(1), 1. https://doi.org/10.1364/josa.61.000001

Lavanya, A. (2025). Real Time Object Detection Using OpenCV. International Journal of Scientific Research in Engineering and Management, 09(06), 1–9. https://doi.org/10.55041/ijsrem51140

Li, M., Liu, J., Yang, W., Sun, X., & Guo, Z. (2018). Structure-Revealing Low-Light Image Enhancement via Robust Retinex Model. Ieee Transactions on Image Processing, 27(6), 2828–2841. https://doi.org/10.1109/tip.2018.2810539

Li, W. (2022). Vehicle Detection in Foggy Weather Based on an Enhanced YOLO Method. Journal of Physics Conference Series, 2284(1), 12015. https://doi.org/10.1088/1742-6596/2284/1/012015

Li, W., Li, S., Wang, Y., & Yun, J. (2022). Study on Personnel Detection Based on Retinex and YOLOv4 in Building Fire. Journal of Physics Conference Series, 2185(1), 12039. https://doi.org/10.1088/1742-6596/2185/1/012039

Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., & Berg, A. C. (2016). SSD: Single Shot MultiBox Detector. 21–37. https://doi.org/10.1007/978-3-319-46448-0_2

Mei, M., Zhou, Z., Liu, W., & Ỹe, Z. (2024). GOI-YOLOv8 Grouping Offset and Isolated GiraffeDet Low-Light Target Detection. Sensors, 24(17), 5787. https://doi.org/10.3390/s24175787

Meng, C., Zhang, J., Chu, H., Xi, K., & Zhao, B. (2022). Remote Sensing Image Enhancement Based on MSR and CLAHE. https://doi.org/10.18178/wcse.2022.04.059

Muhammad, F., Aprilianti, D., Nelvi, A. A., Khairunisa, A., Restu, M., & Kahvi, I. (2024). Perbaikan Kualitas Citra Cahaya Redup Menggunakan Teknik Perbaikan Histogram Equalization dan Adaptive Multi-scale Retinex Low Light Image Enhancement using Histogram Equalization and Adaptive Multi-scale Retinex Repair Techniques. Jurnal Ilmu Komputer Dan Agri-Informatika, 11, 19–26. https://doi.org/10.29244/jika.11.1.19-26

Saadoon, A. M., & Koyuncu, H. (2023). Intelligence Feeder System for Stray Cats. Indonesian Journal of Electrical Engineering and Computer Science, 31(3), 1507. https://doi.org/10.11591/ijeecs.v31.i3.pp1507-1514

Saputra, L. K. P. (2016). Perbandingan Varian Metode Multiscale Retinex untuk Peningkatan Akurasi Deteksi Wajah Adaboost HAAR- HAAR like. 2(April), 89–98.

Xie, C., Tang, H., Fei, L., Zhu, H., & Hu, Y. (2023). IRNet: An Improved Zero-Shot Retinex Network for Low-Light Image Enhancement. Electronics, 12(14), 3162. https://doi.org/10.3390/electronics12143162