EFFD: An Unsupervised Surface Defect Detection Method Based on Estimation and Fusion of Normal Sample Feature Distribution

Train

python main.py
--mode train 
--plus 1 
--device 0 
--batch_size 4 
--num_workers 15 
--image_size 256 256 
--iaff_num_epochs 200 
--iaff_lr 0.00001 
--iaff_weight_decay 0.00005 
--cae_num_epochs 200 
--cae_lr 0.00001 
--cae_weight_decay 0.00005 
--levels level_2_1 level_2_2 level_3_1 level_3_2 level_3_3 level_3_4 level_4_1 level_4_2 level_4_3 level_4_4 
--pool avgpool 
--padding_mode reflect 
--gamma 4 
--alpha 3 
--betas 2 2 2 
--eta 8 8 
--sigma 4 4 
--dataset mvtec 
--categories tile wood cable 
--weights [8,4,1] [8,1,1] [1,4,8] 
--data_path data 
--pretrain_path pretrain 
--evaluate_interval 1

Test

python main.py
--mode test 
--plus 1 
--device 0 
--num_workers 15 
--image_size 256 256 
--levels level_2_1 level_2_2 level_3_1 level_3_2 level_3_3 level_3_4 level_4_1 level_4_2 level_4_3 level_4_4 
--pool avgpool 
--padding_mode reflect 
--gamma 4 
--alpha 3 
--betas 2 2 2 
--eta 8 8 
--sigma 4 4 
--dataset mvtec 
--categories tile wood cable 
--weights [8,4,1] [8,1,1] [1,4,8] 
--data_path data 
--pretrain_path pretrain
--result_path result
--expect_fprs 0.0001 0.0005 0.001