Hello, my name is David Li, I am a student from the University of Michigan that is currently involved in a UAV project sponsored by a local company called Pillar technology. Our goal is to enable farmers to fly drones over their crops(corn crops, for example) and take pictures, then present a seamless picture of the entire crops by stitching individual pictures together, so they could check the conditions of their crops in a faster and low-cost manner. My task right now is to find the best ways to stitch the pictures together. Since this is a sponsored project with limited budget. we prefer free software, but can work with software with reasonable prices. Also, we want the software to stitch geo-referenced rather than simply producing a panoramic picture as pictures for homogeneous crop can be hard to stitch without being geo-tagged. One piece of software we are looking at is an open-source software called OpenDroneMap which runs on Ubuntu. We are going for a solution that works well on Windows, and running a software on virtual machine seems too tedious a process for everyday users. Currently solutions we have in mind: We will try to set up a local web server on our end, and have users interact with a webpage interface where they will upload pictures(retrieved from the SD card of the camera) and receive their resulting picture through email. But we are concerned that since the number of pictures can be huge, the bandwidth requirement maybe high and therefore makes the whole process slow. Using Cygwin to run OpenDroneMap on Windows, we might have to rebuild the software in Windows, and the users will still be required to go through a command line interface. Use an existing Windows image stitching software that can process geo-tagged pictures. This would be straight forward, but we are struggling to find the best one so far. If you have any suggestions or opinions on anything mentioned above, please feel free to post a follow up, and anything will be truly appreciated.