| AFN | Adversarial Fusion Network |
| CBAM | Convolutional Block Attention Module |
| CIoU | Complete Intersection over Union |
| CNN | Convolutional Neural Network |
| DW | Depthwise Convolution |
| Faster R-CNN | Faster Region-based Convolutional Neural Network |
| FN | False Negative |
| FP | False Positive |
| GFLOPs | Giga Floating-Point Operations |
| GELU | Gaussian Error Linear Unit |
| Grad-CAM | Gradient-weighted Class Activation Mapping |
| HSV | Hue, Saturation, and Value |
| LECD | Lightweight Efficient Convolutional Detection head |
| LRM | LECD, RepViTBlock, and MPDIoU (initial letters of the three core modules) |
| mAP | Mean Average Precision |
| mAP50 | Mean Average Precision at IoU = 0.5 |
| mAP50-95 | Mean Average Precision averaged over IoU thresholds from 0.5 to 0.95 |
| MBConv | Mobile Inverted Bottleneck Convolution |
| Mixed-NMS | Mixed Non-Maximum Suppression |
| PW | Pointwise Convolution |
| R-CNN | Region-based Convolutional Neural Network |
| RepViTBlock | Re-parameterized Vision Transformer Block |
| RT-DETR | Real-Time Detection Transformer |
| SE | Squeeze-and-Excitation |
| SPPF | Spatial Pyramid Pooling-Fast |
| SSD | Single Shot MultiBox Detector |
| SVM | Support Vector Machine |
| TP | True Positive |
| UAV | Unmanned Aerial Vehicle |
| YOLO | You Only Look Once |