Journal of Engineering and Applied Science

Table 1 Performance of B-MHAC on TRSE video datasets

From: Train rolling stock video segmentation and classification for bogie part inspection automation: a deep learning approach

Datasets	IoU	mAP	mFI	mNI
B-1	0.9952	0.9639	0.1225	0.142
B-2	0.9057	0.8972	0.2004	0.2222
B-3	0.8987	0.883	0.2554	0.2365
B-4	0.9189	0.9503	0.1509	0.1489
B-5	0.9125	0.9177	0.1838	0.2053
B-6	0.8665	0.7948	0.4325	0.342
Average scores	0.91625	0.90115	0.22425	0.21615

Back to article page