Preparing a Novel Dataset for Improved Suspicious Activity Detection in Academia
Volume 19, Issue 1, 2025
Download| Author(s): |
Asad Hameed Soomro* Shah Abdul Latif University, Khairpur Mirs, Pakistan, asad.soomro31@yahoo.com Rafaqat Hussain Arain Shah Abdul Latif University, Khairpur Mirs, Pakistan, rafaqat.arain@salu.edu.pk Riaz Ahmed Shaikh Shah Abdul Latif University, Khairpur Mirs, Pakistan, riaz.shaikh@salu.edu.pk Vijay Kumar Premy The begum Nusrat Bhutto Women University, Sukkur, Pakistan, vijay.kumar@bnbwu.edu.pk Samar Abbas Mangi Shah Abdul Latif University, Khairpur Mirs, Pakistan, mangisamar@gmail.com Waseem Ahmed Buriro Sukkur IBA University, Sukkur, Pakistan, burirowaseemahmed@gmail.com |
|---|---|
| Abstract | Human Activity Recognition (HAR) is one of the important research areas with essential security applications in a variety of environments, including academic environments. However, available datasets frequently lack the real-world diversity required for effective training of HAR models. To bridge this gap, this paper presents the CampusWatch dataset, specifically collected to capture real-world activities within academic institutions for Suspicious Activity Detection (SAD). The dataset, collected through smartphone cameras, spans both indoor and outdoor scenes from academia. It includes ten activity categories, covering nine suspicious behaviors—such as kicking, punching, running—and tenth one is "normal" activity class. This paper also outlines key steps in data preparation, including data collection, challenges, and techniques such as video annotation, noise reduction, and preprocessing, providing a robust foundation for advancing research in SAD. |
| Keywords | Human Activity Recognition; Suspicious Activity Detection; security applications; HAR models; academic institutions; THUMOS. |
| Year | 2025 |
| Volume | 19 |
| Issue | 1 |
| Type | Research paper, manuscript, article |
| Journal Name | Journal of Information & Communication Technology | Publisher Name | ILMA University | Jel Classification | - | DOI | - | ISSN no (E, Electronic) | 2075-7239 | ISSN no (P, Print) | 2415-0169 | Country | Pakistan | City | Karachi | Institution Type | University | Journal Type | Open Access | Manuscript Processing | Blind Peer Reviewed | Format | Paper Link | https://jict.ilmauniversity.edu.pk/journal/jict/19.1/2.pdf | Page | 9-13 |