An Improved Deep Residual Network-Based Semantic Simultaneous Localization and Mapping Method for Monocular Vision Robot

Table 2

The experimental results of semantic recognition in the mix environment.

Semantic labels	Total number of scenes	False identification of scenes	Recognition accuracy (%)
West	80	10	87.50
South	80	6	92.50
East	80	5	93.50
North	80	8	90.00
Inner	80	0	100.00
Average value	80	5.8	92.70