Development and characterization of a dimmable LED luminaire for body detection, tracking and recognition under natural and artificial low-light illumination

• Main Authors: Llorente, Cesar A., Dadios, Elmer P.
• Format: text
• Published: Animo Repository 2019
• Subjects: LED lighting; Cameras—Lighting; Motion detectors; Electrical and Computer Engineering; Electrical and Electronics
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/2447; https://animorepository.dlsu.edu.ph/context/faculty_research/article/3446/type/native/viewcontent
• Institution: De La Salle University

Illumination of the scene is crucial in order the detection algorithm to carry out the computation task correctly. In this study, an LED based luminaire is developed for a computer vision system for the detection and tracking of pedestrian or human body in a video under different illumination. The illumination is provided using an LED lighting system where the intensity of illumination can be varied by changing the duty cycle of the pulse activating the LEDs. The mean of ambient light intensities captured by the image sensor is captured under different lighting conditions which will be useful in designing a control system that consider the amount of ambient lighting to set the setting of LED lighting activation. This adaptive system is useful to provide illumination that take into consideration ambient light intensities. © 2019, World Academy of Research in Science and Engineering. All rights reserved.