ABi-LSTM |
Attention enhanced bidirectional long short-Term memory network |
AI |
Artificial intelligence |
AP |
Average precision |
BiFPN |
Bidirectional feature pyramid network |
CAM |
Channel attention module |
CBAM |
Convolutional block attention module |
CBS |
Convolution, batch normalization, SiLU activation function |
CIoU |
Complete intersection over union |
CNN |
Convolutional neural network |
CSP |
Convolutional spatial pyramid |
DOTA |
Dataset for object detection in aerial images |
DP |
Decoupled head |
ECA |
Efficient channel attention |
E-ELAN |
Extended efficient layer aggregation network |
FLAME |
Fire luminosity airborne-based machine learning evaluation |
FLOPS |
Floating point operations per second |
FPN |
Feature pyramid network |
FP |
False positive |
FPS |
Frame per second |
FN |
False negative |
GAP |
Global average pooling |
GB |
Gigabyte |
GPU |
Graphics processing unit |
GFLOPS |
GPU floating point operations per second |
GHz |
Gigahertz |
GMP |
Global max pooling |
GIoU |
Generalized intersection over union |
GUI |
Graphical user interface |
IoU |
Intersection over union |
Microsoft COCO |
Common objects in context |
NEMO |
Nevada smoke detection benchmark |
PANet |
Path aggregation network |
ReLU |
Rectified linear unit |
SAM |
Spatial attention module |
SE |
Squeeze-and-excitation |
SiLU |
Sigmoid linear unit |
SPPCSPC |
Spatial pyramid pooling cross stage partial network |
SPPF+ |
Spatial pyramid pooling fast+ |
SSD |
Single-shot detector |
TN |
True negative |
TP |
True positive |
UAV |
Unmanned aerial vehicle |
VGG16 |
Visual geometry group 16 |
YOLO |
You only look once |