Research Article
A Lightweight Dangerous Liquid Detection Method Based on Depthwise Separable Convolution for X-Ray Security Inspection
Algorithm 1
Semiautomatic labeling method.
| Input: The dual-energy X-ray dataset, the lightweight object location network and the confidence score threshold T | | Output: The dual-energy X-ray dataset with labels and the trained lightweight object location network | (1) Train with Ul | (2) | (3) For i = 1, , i++ do predict with , get confidence score and Location Li; | (4) Fordo | (5) If or manually verify the existence of undetected object then | (6) Remove into ; | (7) Delete the corresponding object information; | (8) Else | (9) Update the labels of ; | (10) End if | (11) End for | (12) Label manually; | (13) ; | (14) Fine-tune with ; | (15) Empty ; | (16) End for | (17) | Return and ; |
|