Proposed Algorithm. by Hend Bayoumi (22693738)

Subjects:

Test results of different models on TinyPerson. by Ju Liang (4277062)

“…These characteristics impose high demands on detection algorithms in terms of fine-grained feature extraction, cross-scale fusion capability, and occlusion resistance.The YOLOv11s model has significant limitations in practical applications: its feature extraction module has a single semantic representation, the traditional feature pyramid network has limited capability to detect multi-scale targets, and it lacks an effective feature compensation mechanism when targets are occluded.To address these issues, we propose a UAV aerial small target detection algorithm named UAS-YOLO (Universal Inverted Bottleneck with Adaptive BiFPN and Separated and Enhancement Attention module YOLO), which incorporates three key optimizations. …”

Comparisons between ADAM and NADAM optimizers. by Hend Bayoumi (22693738)

Subjects:

Visual comparison of TinyPerson. by Ju Liang (4277062)

“…These characteristics impose high demands on detection algorithms in terms of fine-grained feature extraction, cross-scale fusion capability, and occlusion resistance.The YOLOv11s model has significant limitations in practical applications: its feature extraction module has a single semantic representation, the traditional feature pyramid network has limited capability to detect multi-scale targets, and it lacks an effective feature compensation mechanism when targets are occluded.To address these issues, we propose a UAV aerial small target detection algorithm named UAS-YOLO (Universal Inverted Bottleneck with Adaptive BiFPN and Separated and Enhancement Attention module YOLO), which incorporates three key optimizations. …”

Comparisons of computation rate performance for different offloading algorithms.for N = 10, 20, 30. by Hend Bayoumi (22693738)

Subjects:

Comparison of total time consumed for different offloading algorithms.for N = 10, 20, 30. by Hend Bayoumi (22693738)

Subjects:

An Example of a WPT-MEC Network. by Hend Bayoumi (22693738)

Subjects:

Related Work Summary. by Hend Bayoumi (22693738)

Subjects:

Simulation parameters. by Hend Bayoumi (22693738)

Subjects:

Training losses for N = 10. by Hend Bayoumi (22693738)

Subjects:

Performance Comparison of State-of-the-Art WPT-MEC Offloading Methods. by Hend Bayoumi (22693738)

Subjects:

The evolution of the Wireless Power Transfer (WPT) time fraction β over simulation frames. by Hend Bayoumi (22693738)

Subjects:

Normalized computation rate for N = 10. by Hend Bayoumi (22693738)

Subjects:

CDF of task latency, approximated as the inverse of the achieved computation rate. by Hend Bayoumi (22693738)

Subjects:

Comparison between Bisection and Newton-Raphson in terms of (A) total time consumed, and (B) maximum computation rate. by Hend Bayoumi (22693738)

Subjects:

Summary of Notations Used in this paper. by Hend Bayoumi (22693738)

Subjects:

Test results of different models on VisDrone2019. by Ju Liang (4277062)

“…These characteristics impose high demands on detection algorithms in terms of fine-grained feature extraction, cross-scale fusion capability, and occlusion resistance.The YOLOv11s model has significant limitations in practical applications: its feature extraction module has a single semantic representation, the traditional feature pyramid network has limited capability to detect multi-scale targets, and it lacks an effective feature compensation mechanism when targets are occluded.To address these issues, we propose a UAV aerial small target detection algorithm named UAS-YOLO (Universal Inverted Bottleneck with Adaptive BiFPN and Separated and Enhancement Attention module YOLO), which incorporates three key optimizations. …”

Structure of YOLOv11 network. by Ju Liang (4277062)

“…These characteristics impose high demands on detection algorithms in terms of fine-grained feature extraction, cross-scale fusion capability, and occlusion resistance.The YOLOv11s model has significant limitations in practical applications: its feature extraction module has a single semantic representation, the traditional feature pyramid network has limited capability to detect multi-scale targets, and it lacks an effective feature compensation mechanism when targets are occluded.To address these issues, we propose a UAV aerial small target detection algorithm named UAS-YOLO (Universal Inverted Bottleneck with Adaptive BiFPN and Separated and Enhancement Attention module YOLO), which incorporates three key optimizations. …”

Structure of SEAM network [16]. by Ju Liang (4277062)

“…These characteristics impose high demands on detection algorithms in terms of fine-grained feature extraction, cross-scale fusion capability, and occlusion resistance.The YOLOv11s model has significant limitations in practical applications: its feature extraction module has a single semantic representation, the traditional feature pyramid network has limited capability to detect multi-scale targets, and it lacks an effective feature compensation mechanism when targets are occluded.To address these issues, we propose a UAV aerial small target detection algorithm named UAS-YOLO (Universal Inverted Bottleneck with Adaptive BiFPN and Separated and Enhancement Attention module YOLO), which incorporates three key optimizations. …”

Ablation experiment curve. by Ju Liang (4277062)

“…These characteristics impose high demands on detection algorithms in terms of fine-grained feature extraction, cross-scale fusion capability, and occlusion resistance.The YOLOv11s model has significant limitations in practical applications: its feature extraction module has a single semantic representation, the traditional feature pyramid network has limited capability to detect multi-scale targets, and it lacks an effective feature compensation mechanism when targets are occluded.To address these issues, we propose a UAV aerial small target detection algorithm named UAS-YOLO (Universal Inverted Bottleneck with Adaptive BiFPN and Separated and Enhancement Attention module YOLO), which incorporates three key optimizations. …”