EXP-DDQN algorithm. by Sen Cao (6017846)

“…<div><p>With the increasing integration of Connected and Automated Vehicles (CAVs) and Human-Driven Vehicles (HDVs) in urban traffic systems, along with highly variable pedestrian crossing demands, traffic management faces unprecedented challenges. …”

Deep reinforcement learning process. by Sen Cao (6017846)

“…<div><p>With the increasing integration of Connected and Automated Vehicles (CAVs) and Human-Driven Vehicles (HDVs) in urban traffic systems, along with highly variable pedestrian crossing demands, traffic management faces unprecedented challenges. …”

Simulated road network. by Sen Cao (6017846)

“…<div><p>With the increasing integration of Connected and Automated Vehicles (CAVs) and Human-Driven Vehicles (HDVs) in urban traffic systems, along with highly variable pedestrian crossing demands, traffic management faces unprecedented challenges. …”

EXP-DDQN algorithm framework. by Sen Cao (6017846)

“…<div><p>With the increasing integration of Connected and Automated Vehicles (CAVs) and Human-Driven Vehicles (HDVs) in urban traffic systems, along with highly variable pedestrian crossing demands, traffic management faces unprecedented challenges. …”

Comparison of metrics across different methods. by Sen Cao (6017846)

“…<div><p>With the increasing integration of Connected and Automated Vehicles (CAVs) and Human-Driven Vehicles (HDVs) in urban traffic systems, along with highly variable pedestrian crossing demands, traffic management faces unprecedented challenges. …”

Phage–host adsorption results. by Rokaiya Nurani Shatadru (22656316)

Structure diagram of DCNv3. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Ablation experiment. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

PR comparison. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Flexi-YOLO structure. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Experimental environment. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Loss curve. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Results of the inference experiment. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Distribution of dataset annotations. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

C3 Module. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Variable convolution process. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

GAM structure diagram. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Hyperparameter settings. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Ghost detect module. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”

Example of preprocessed image. by Jiexiang Yang (19980594)

“…Experimental results show that Flexi-YOLO achieves an accuracy increase of 2.7% over YOLOv8n, a recall rate rise of 4.7%, a mAP improvement of 5.3%, a mAP@0.5–0.95 increase of 3.9%, a decrease of 0.5 in GFLOPS, and an F1 score improvement from 0.80 to 0.84. …”