Precise Detection of Changes in Relatively Static Environments for Single-Board Computers

This paper proposes a computationally efficient change detection algorithm designed for resource-constrained devices, such as single-board computers (e.g., Raspberry Pi), for monitoring changes in relatively static environments like museums. The proposed method combines a color adjustment technique...

Full description

Saved in:
Bibliographic Details
Published in2023 14th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI) pp. 57 - 62
Main Authors Nishidate, Yohei, Kohira, Yukihide, Takahashi, Shigeo, Yoshioka, Rentaro
Format Conference Proceeding
LanguageEnglish
Published IEEE 08.07.2023
Subjects
Change detection
Change detection algorithms
Computers
Edge devices
Histograms
Human-centric systems
Image color analysis
Image registration
Multi-temporal images
Remote surveillance
Surveillance
Visualization
Online AccessGet full text

Cover

More Information
Summary:This paper proposes a computationally efficient change detection algorithm designed for resource-constrained devices, such as single-board computers (e.g., Raspberry Pi), for monitoring changes in relatively static environments like museums. The proposed method combines a color adjustment technique using Lab color scale conversion and Contrast Limited Adaptive Histogram Equalization (CLAHE), image registration by Enhanced Correlation Coefficient (ECC) maximization with homography model, and a simple thresholding approach for change detection. The effectiveness of the algorithm is assessed through four different situations in the Fukushima prefectural museum. The proposed method shows promise as a practical and efficient solution for resource-constrained devices and has the potential to contribute significantly to enhanced assistance of surveillance and monitoring across a wide range of applications. However, some limitations and detection failures remain, which could be addressed in future work by improving the color adjustment method and further reducing the impact of artifacts in the change detection process.
DOI:10.1109/IIAI-AAI59060.2023.00022