Visualization technologies are widely used in various fields. However, they are often restricted by environmental characteristics. Therefore, this paper presents a development of a hybrid mixed reality (MR) visualization system that enables high precision superimposition in any environment using location-based and marker-based methods. A Microsoft HoloLens 2 was used in the MR device to visualize and QZNEO receivers were used to receive global navigation satellite system (GNSS) data. This system can maintain high precision superimposition by switching between the superimposition methods according to the reception level of location information. The location-based and marker-based methods were used in environment where the receivers exhibited good and poor reception for location information, respectively. The proposed system was used to visualize underwater objects. The results demonstrated the effectiveness and validity of the proposed system.