This project presents an indoor navigation system based on image matching, aiming to address the challenges of localization and navigation in indoor environments. The system utilizes Simultaneous Localization and Mapping (SLAM) technology to capture high-resolution images and point cloud data, combined with the VGG16 model from Convolutional Neural Networks (CNN) for image processing, feature extraction, and matching.

In our research, we conducted experiments at the Faculty of Architecture and the Built Environment of Delft University of Technology, using a SLAM scanner to obtain 360-degree panoramic images and point cloud data of the indoor environment. Through cube mapping projection, we converted the panoramic images into six planar views, selecting the front, right, and left views as positioning references. Additionally, we reconstructed the indoor environment structure and designed node networks for positioning and navigation.

The technical architecture of this system comprises three main components: VGG16-based image feature extraction, cosine similarity-based image matching, and DBSCAN algorithm for location clustering. Through this method, the system can achieve real-time localization results after image capture and provide users with optimal paths using the A* navigation algorithm.

Experimental results show that when using single image matching, the system's room localization accuracy reaches 74.65\%. When employing multiple image matching and DBSCAN clustering methods, the accuracy significantly improves. In our final evaluation involving 116 positions, the system successfully matched 111 of these positions to their correct rooms, achieving a localization accuracy of 95.69\%.

This research not only provides an innovative solution for indoor positioning and navigation but also points the way for future research, including support for multi-floor navigation, enhancing CNN model performance, and automating building processing. This technology has the potential for widespread application in complex indoor environments such as large buildings, conference centers, and university campuses, offering users accurate, real-time positioning and navigation services.

Due to safety or preservation reasons, certain objects or areas need to be inspected regularly. Currently, the inspection of objects is mostly done in-person, which is labour-intensive and not very effective. With the use of drones, areas and objects can be inspected from new angles at a much faster rate. To effectively monitor these objects with drones, the position and location needs to be extracted from drone data in real time.
In this thesis, a case study is done on the localisation of fisher boats in restricted areas. Several components are integrated to create a prototype. The pretrained YOLOv3 detection model is trained on acquired nadir boat images, which makes it able to predict the bounding boxes of boats on images captured with drones. A positioning algorithm is constructed, which calculates the geographical coordinates from the pixel coordinates for images taken both in a nadir and an oblique angle. A real time connection is constructed between the drone and the prototype. This is done by creating a connection with Google Drive with the drone controller and the prototype. The positioned polygon bounding boxes are localised using a real time dashboard, which visualises the bounding boxes in a map with other relevant layers.
The results indicate that the performance of the components and prototype as a whole are satisfactory for this use case. To deploy this prototype in other object localisation use cases, it is recommended to train the pretrained model further, use a drone with more accurate equipment and run the prototype on the drone controller.