Deep-Learning-Approach-for-Surface-Defect-Detection
Deep-Learning-Approach-for-Surface-Defect-Detection copied to clipboard
ShuaiLYU
How many epochs for segment and decision net training and is the output of segment network needs to be binarized？
The numbers of epochs is not fixed. I didn't binarized the output of segment network, but you can have of try. Maybe, you can get a better result.
The numbers of epochs is not fixed. I didn't binarized the output of segment network, but you can have of try. Maybe, you can get a better result.
Hello! I am puzzled with my test results. I use the same dataset as you and followed your instructions.However, my results were still very different from the results you given in the folder ./visualization/test(where the output of the segmentation network have been binarized and the defects are wonderfully detected). In my test result, not only were the output of the segmentation results not binarized, but also it seems hard to find where the defects are from the output of the segmentation network even with my eyes. I wonder how can I get the result like you given in the folder ./visualization/test? Thank you very much!
Hello! I am puzzled with my test results. I use the same dataset as you and followed your instructions.However, my results were still very different from the results you given in the folder ./visualization/test(where the output of the segmentation network have been binarized and the defects are wonderfully detected). In my test result, not only were the output of the segmentation results not binarized, but also it seems hard to find where the defects are from the output of the segmentation network even with my eyes. I wonder how can I get the result like you given in the folder ./visualization/test? Thank you very much!
oh, me to. I am puzzled too
