Indoor scene recognition method and system based on combined semantic region relation model

The invention discloses an indoor scene recognition method and system based on a combined semantic region relation model. The method comprises the following steps: acquiring an indoor scene image to be recognized; inputting the obtained image into a combined semantic region relation model, extracting semantic global node features of the input image through a semantic region relation module, extracting global node features of the input image through an RGB feature extraction module, and aggregating the semantic global node features and the global node features through a global feature aggregation module to obtain a semantic global node feature set; and global aggregation node features are obtained, the extracted global aggregation node features are input to a classifier, and a scene recognition result is output. According to the method, a semantic region relation model is constructed, the relation between object semantic regions in an indoor scene is fully mined, and the semantic segmentation error problem in