As I wrote to you in the private email, the connectivity image reveals some longitudinal white gaps that indicate weak matching of images.
The Pix4D orthomosaic shows a quite homogeneous sandy terrain, which will lead to difficulties in extracting common characteristic points (keypoints) and thus explains the weak matching. On the other hand, the image from Google Earth looks much better, but this is probably due to the large size of the image swath (=the area imaged on the surface by a satellite sensor; image swaths usually range from tens to hundreds of km width), which is to be expected in the case of satellite images. If a swath is very large, then there is no stitching involved, as one single image can already cover the area of interest. Google Earth is exclusively comprised of satellite images and aerial photography (which are also not updated).
If it is possible to fly again, I would recommend the following:
- Use accurate image geolocation with DJI Phantom 4 RTK (P4RTK).
- Increase the overlap between images to at least 85% frontal overlap and at least 70% side overlap.
- Fly higher. In most cases, flying higher improves the results.
Regarding fine-tuning the project parameters, I would recommend:
Hopefully this helps.