A computer vision project using YOLO26 for real-time human detection in search and rescue scenarios. Trained and compared two model variants (Nano and Medium), with the best model achieving 97.24% mAP50 and 72.46% mAP50-95 on the test set.
Note: This project uses the Ultralytics framework with yolo26n.pt and yolo26m.pt pretrained weights.
Problem Statement
In search and rescue operations, rapid identification of humans in diverse environments (aerial footage, disaster zones, wilderness) is critical. Manual review of imagery is time-consuming and prone to errors under pressure. This project develops a deep learning solution to:
Detect humans accurately in challenging conditions (varied lighting, occlusion, terrain)
Provide real-time inference suitable for drone/UAV deployment
Minimize false negatives to ensure no person is missed
Class Loss: Classification error (human vs. background)
DFL Loss: Distribution Focal Loss for precise box boundaries
Results
Model Performance Comparison
YOLO26 Nano:
Split
Precision
Recall
mAP50
mAP50-95
Validation
96.06%
90.18%
96.34%
68.05%
Test
92.98%
92.18%
96.40%
67.93%
YOLO26 Medium (Best Model):
Split
Precision
Recall
mAP50
mAP50-95
Validation
96.73%
94.03%
97.43%
72.77%
Test
94.31%
94.38%
97.24%
72.46%
Performance Improvement (Medium vs Nano on Test Set):
Precision: +1.33%
Recall: +2.20%
mAP50: +0.84%
mAP50-95: +4.53%
Interpretation:
YOLO26 Medium provides the best overall performance with 94.31% precision and 94.38% recall
High recall (94.38%) ensures most humans are detected - critical for SAR where missing a person has serious consequences
High precision (94.31%) means low false positives (~5.7%), reducing wasted effort investigating false alarms
97.24% mAP50 demonstrates strong localization accuracy at standard IoU threshold
72.46% mAP50-95 shows good performance even with strict overlap requirements
Medium model's 4.53 point improvement in mAP50-95 indicates significantly better bounding box precision
Training Progression (Medium)
Training Curve Medium
Key Observations:
Losses steadily decreased and plateaued around epoch 150-200
Validation metrics stabilized with minimal overfitting
Early stopping triggered at epoch 243 (no improvement for N epochs)
Qualitative Results (Medium)
Sample Detections:
Training Curve Medium
Comparison between Nano and Medium:
Comparison
Observation: Both model successfully detects partially occluded humans. Medium model consistently performs better than nano.
Key Insights
From Model Comparison (Nano vs Medium):
Model Size vs Performance Trade-off: Medium model achieves 4.53% in mAP50-95 over Nano, indicating significantly better localization precision
Recall Improvement: Medium model's +2.2% recall gain means detecting ~2% more humans - crucial for SAR
Balanced Performance: Medium model achieves balanced precision (94.31%) and recall (94.38%), avoiding the precision-recall trade-off
Deployment Consideration: Nano model offers 96.40% mAP50 with faster inference - viable for resource-constrained drones if speed is prioritized
From Training Process:
Early Stopping Effectiveness: Training stopped at epoch 243/300, preventing overfitting while achieving strong generalization
Transfer Learning Success: Pretrained YOLO26 weights accelerated convergence and improved final performance
Stable Training: Cosine LR decay with AdamW optimizer enabled smooth convergence without oscillations
From Model Performance:
High Recall Priority: 94.38% recall ensures minimal missed detections - critical for SAR where missing a person has serious consequences
Low False Positive Rate: 94.31% precision means only ~5.7% false positives, reducing wasted effort investigating false alarms
Robust Localization: 72.46% mAP50-95 indicates bounding boxes are well-calibrated across strict IoU thresholds
Strong Generalization: Test metrics (97.24% mAP50) closely match validation (97.43%), confirming model isn't overfitting
Deployment Considerations:
Real-Time Capable: YOLO26 architecture enables 152.3 FPS on NVIDIA L40S, suitable for drone inference
ONNX Export: Both models successfully exported to ONNX for cross-platform deployment (edge devices, web, mobile)
Model Selection: Use Medium for maximum accuracy, Nano for speed-critical applications
Failure Modes: Struggle with extremely small objects, and tends to perform worse on direct top down view where the leg and hand may be hidden by the head.
Future Improvements
1. Multi-Class Detection (If Applicable)
Extend to detect specific categories: injured persons, rescue personnel, animals
Enable priority tagging for triage in mass casualty events
2. Temporal Consistency (for video SAR missions)
Implement tracking (DeepSORT, ByteTrack) to associate detections across frames
Count unique individuals rather than per-frame detections
Reduce false positives by filtering isolated single-frame detections
3. Adversarial Robustness
Test on extreme weather conditions (fog, heavy rain, snow)
Augment training data with synthetic weather overlays
Evaluate performance degradation and set realistic confidence thresholds
4. Uncertainty Quantification
Implement Monte Carlo Dropout or ensemble methods for confidence estimation
