Figures & data
Figure 1. The visualisation results of the training dataset by Labelme annotation tool, presenting original images (a, c) and labelled images (b, d) of certain samples.
![Figure 1. The visualisation results of the training dataset by Labelme annotation tool, presenting original images (a, c) and labelled images (b, d) of certain samples.](/cms/asset/2994fd44-40cb-46c3-8ea1-b3d35b098c91/tjde_a_2384631_f0001_oc.jpg)
Figure 2. The DeepLabv3 + network backbone structure is usinged in both semi-supervised (UniMatch) and fully supervised models.
![Figure 2. The DeepLabv3 + network backbone structure is usinged in both semi-supervised (UniMatch) and fully supervised models.](/cms/asset/e2943fab-a8ef-4b04-9680-54443d909928/tjde_a_2384631_f0002_oc.jpg)
Figure 3. Architecture of fully supervised framework (a) and semi-supervised framework (UniMatch) (b).
![Figure 3. Architecture of fully supervised framework (a) and semi-supervised framework (UniMatch) (b).](/cms/asset/82b1d32e-f4b1-47d1-b7db-baf0977065b0/tjde_a_2384631_f0003_oc.jpg)
Figure 4. Numerical variation diagram of loss values in the supervised natural road training process (a) and semi-supervised natural road training process (b).
![Figure 4. Numerical variation diagram of loss values in the supervised natural road training process (a) and semi-supervised natural road training process (b).](/cms/asset/fe29218a-c3d8-4b0c-906b-54a8b765e060/tjde_a_2384631_f0004_oc.jpg)
Figure 5. Numerical variation diagram of IOU and MIOU values in the supervised resnet101 -based DeepLabv3 training process (a,c) and semi-supervised UniMatch training process with 1008 unlabelled images (b,d).
![Figure 5. Numerical variation diagram of IOU and MIOU values in the supervised resnet101 -based DeepLabv3 training process (a,c) and semi-supervised UniMatch training process with 1008 unlabelled images (b,d).](/cms/asset/e6fed5a9-6d50-4ed0-8903-1f23decb9818/tjde_a_2384631_f0005_oc.jpg)
Figure 6. Several sample recognition result images were obtained using the training model of UniMatch semi-supervised network (a, b).
![Figure 6. Several sample recognition result images were obtained using the training model of UniMatch semi-supervised network (a, b).](/cms/asset/f7e0a756-573c-4ad2-a22a-d41d08e9f0f0/tjde_a_2384631_f0006_oc.jpg)
Table 1. Best evaluation indexes of generated model files by supervised and UniMatch semi-supervised network.
Data availability statement
Data and model have been made available on the following GitHub link: https://github.com/wm-jessica/roadseg-back-up.