Intelligent Identification of Coal Crack in CT Images Based on Deep Learning

Figure 4

Modified DeepLabV3+ model structure. Compared with the original DeepLabV3+ architecture, the OS (output stride) was adjusted to 8, and rates of the ASPP module were adjusted to (1, 12, 24, 36). At the same time, the first upsampling rate is changed from 4 to 2 to restore image pixels to their original size.