The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera Selection
Multi-camera target tracking is a critical technology for continuous monitoring in large-scale environments, with applications in smart cities, security surveillance, and emergency response. However, existing tracking systems often suffer from high computational costs and energy inefficiencies, part...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/6/3052 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849392720780460032 |
|---|---|
| author | Jaemin Yang Jongwoo Lee Ilju Lee Yaesop Lee |
| author_facet | Jaemin Yang Jongwoo Lee Ilju Lee Yaesop Lee |
| author_sort | Jaemin Yang |
| collection | DOAJ |
| description | Multi-camera target tracking is a critical technology for continuous monitoring in large-scale environments, with applications in smart cities, security surveillance, and emergency response. However, existing tracking systems often suffer from high computational costs and energy inefficiencies, particularly in resource-constrained edge computing environments. Traditional methods typically rely on static or heuristic-based camera selection, leading to redundant computations and suboptimal resource allocation. This paper introduces a novel framework for efficient single-target tracking using edge-based distributed smart cameras with dynamic camera selection. The proposed framework employs context-aware dynamic camera selection, activating only the cameras most likely to detect the target based on its predicted trajectory. This approach is designed for resource-constrained environments and significantly reduces computational load and energy consumption while maintaining high tracking accuracy. The framework was evaluated through two experiments. In the first, single-person tracking was conducted across multiple routes with various target behaviors, demonstrating the framework’s effectiveness in optimizing resource utilization. In the second, the framework was applied to a simulated urban traffic light adjustment system for emergency vehicles, achieving significant reductions in computational load while maintaining equivalent tracking accuracy compared to an always-on camera system. These findings highlight the robustness, scalability, and energy efficiency of the framework in edge-based camera networks. Furthermore, the framework enables future advancements in dynamic resource management and scalable tracking technologies. |
| format | Article |
| id | doaj-art-b2a4c4e0387d46c8bf5f6f7c5f7f5fdb |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-b2a4c4e0387d46c8bf5f6f7c5f7f5fdb2025-08-20T03:40:43ZengMDPI AGApplied Sciences2076-34172025-03-01156305210.3390/app15063052The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera SelectionJaemin Yang0Jongwoo Lee1Ilju Lee2Yaesop Lee3Department of Robotics, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Robotics, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Robotics, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Robotics, Kwangwoon University, Seoul 01897, Republic of KoreaMulti-camera target tracking is a critical technology for continuous monitoring in large-scale environments, with applications in smart cities, security surveillance, and emergency response. However, existing tracking systems often suffer from high computational costs and energy inefficiencies, particularly in resource-constrained edge computing environments. Traditional methods typically rely on static or heuristic-based camera selection, leading to redundant computations and suboptimal resource allocation. This paper introduces a novel framework for efficient single-target tracking using edge-based distributed smart cameras with dynamic camera selection. The proposed framework employs context-aware dynamic camera selection, activating only the cameras most likely to detect the target based on its predicted trajectory. This approach is designed for resource-constrained environments and significantly reduces computational load and energy consumption while maintaining high tracking accuracy. The framework was evaluated through two experiments. In the first, single-person tracking was conducted across multiple routes with various target behaviors, demonstrating the framework’s effectiveness in optimizing resource utilization. In the second, the framework was applied to a simulated urban traffic light adjustment system for emergency vehicles, achieving significant reductions in computational load while maintaining equivalent tracking accuracy compared to an always-on camera system. These findings highlight the robustness, scalability, and energy efficiency of the framework in edge-based camera networks. Furthermore, the framework enables future advancements in dynamic resource management and scalable tracking technologies.https://www.mdpi.com/2076-3417/15/6/3052target trackingedge computingmulti-camera systemdynamic camera selectionIoT |
| spellingShingle | Jaemin Yang Jongwoo Lee Ilju Lee Yaesop Lee The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera Selection Applied Sciences target tracking edge computing multi-camera system dynamic camera selection IoT |
| title | The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera Selection |
| title_full | The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera Selection |
| title_fullStr | The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera Selection |
| title_full_unstemmed | The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera Selection |
| title_short | The TEDDY Framework: An Efficient Framework for Target Tracking Using Edge-Based Distributed Smart Cameras with Dynamic Camera Selection |
| title_sort | teddy framework an efficient framework for target tracking using edge based distributed smart cameras with dynamic camera selection |
| topic | target tracking edge computing multi-camera system dynamic camera selection IoT |
| url | https://www.mdpi.com/2076-3417/15/6/3052 |
| work_keys_str_mv | AT jaeminyang theteddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection AT jongwoolee theteddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection AT iljulee theteddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection AT yaesoplee theteddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection AT jaeminyang teddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection AT jongwoolee teddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection AT iljulee teddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection AT yaesoplee teddyframeworkanefficientframeworkfortargettrackingusingedgebaseddistributedsmartcameraswithdynamiccameraselection |