Authors :
Parashurama H.; M. M. Harshitha; Dr. Girish Kumar D.
Volume/Issue :
Volume 11 - 2026, Issue 5 - May
Google Scholar :
https://tinyurl.com/3zfau7ks
Scribd :
https://tinyurl.com/mryxa57k
DOI :
https://doi.org/10.38124/ijisrt/26May890
Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.
Abstract :
Darklytic is developed to address the challenges of identifying objects in night-time environments. Traditional
vision-based detection systems often fail under poor illumination due to noise, low contrast, and loss of visual details. To
overcome these limitations, the proposed system integrates infrared-assisted imaging with an optimized deep learning-based
object detection model. By enhancing grayscale image features and applying efficient preprocessing techniques, Darklytic
improves object visibility and detection accuracy even in near-dark conditions.
References :
- G. Jocher, A. Chaurasia, and J. Qiu, “Ultralytics YOLOv8,” Jan. 2023. [Online]. Available: https://github.com/ultralytics/ultralytics
- Espressif Systems, “ESP32-CAM hardware design and pin mapping,” Espressif Technical Documentation, 2021. [Online]. Available: https://www.espressif.com
- J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, real-time object detection,” in Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779–788.
- S. Patel and M. Hussain, “Real-time video surveillance using ESP32-CAM and edge computing,” Int. J. of Internet of Things and Cloud Computing, 45–52, 2022.
- G. Bradski, “The OpenCV Library,” Dr. Dobb's Journal of Software Tools, vol. 25, pp. 120–123, 2000.
- A. Bochkovskiy, C. Y. Wang, and H. Y. Liao, “YOLOv4: detection,” arXiv preprint arXiv:2004.10934, 2020.
- L. Wang and R. Gupta, “Low-latency video streaming protocols for ESP32-based IoT devices,” IEEE Trans. on Consumer Electronics, vol. 67, no. 2, pp. 112–120, 2021.
- R. Sharma and V. Mehta, “Implementation of human detection algorithms on edge devices,” Journal of Real-Time Image Processing, vol. 18, no. 4, pp. 1245–1259, 2021.
- AI-Thinker Technology, “ESP32-CAM product specification and camera driver integration,” AI-Thinker Open Docs, 2019.
- S. Kumar and J. Park, “Deep learning for night vision: A study on low-light object detection,” Int. J. of Computer Vision and Artificial Intelligence, vol. 12, no. 1, pp. 33–42, 2023.
Darklytic is developed to address the challenges of identifying objects in night-time environments. Traditional
vision-based detection systems often fail under poor illumination due to noise, low contrast, and loss of visual details. To
overcome these limitations, the proposed system integrates infrared-assisted imaging with an optimized deep learning-based
object detection model. By enhancing grayscale image features and applying efficient preprocessing techniques, Darklytic
improves object visibility and detection accuracy even in near-dark conditions.