Stitching generates a georeferenced image overlay. This can be very helpful to detect potential errors in georeferencing. This can be run on a laptop in the field for direct control of the dataset. This option requires georeferencing based on an Ardupilot dataflash log file.
To reduce the parallax effects the images are clipped. The clipping value can be adjusted in case there are some gaps. A misalignment of the camera or a wrong calibration of a NADIR gimbal can easily be detected.
To adjust the yaw offset and a potential timing offset you have to run the geotagging process again with adjusted values. Then you can re-run stitching to observe the effect. This process can be repeated as often as required. However, best results can only be obtained with nadir images.
When using Ardupilot dataflash *.log files for georeferencing information on pitch roll and yaw can be used for stitching to rectify the images accordingly. This model assumes a flat surface and does not account for lens distortion. However, in many cases it can provide information for decision making whithin minutes after finishing the mission in the field.
When working with Agrowing data, Mavis automatically stitches the RGN and the NGN images and generates exactly overlaying Photomosaics.