Research on Indoor Positioning Method Based on Improved HS-AlexNet Model

Scene recognition is the key to achieving accurate and fast indoor positioning. Deep network has become a research central issue lately with its outstanding performance. This paper puts forward an advanced AlexNet network model combined with Harris feature detection, which guides the image processing of the original model according to the detected Harris characteristics, it reduces the randomness error and improves the generalization ability and robustness of the model. In addition, for the campus indoor scene positioning environment, the original structure of the AlexNet network model and the data augmentation method are improved, so that the positioning model can cope with the complex and variable positioning environment, and its accuracy and speed reach a high level. The method can be combined with the existing mainstream visual indoor positioning method to enhance the accurateness and speed of positioning system.