Adding and aligning/stitching photos
This stage entails loading the UAV images into your processing software and stitching (georeferencing) them into a seamless image. Normally, information from the GPS device fitted with the UAV (also known as geotags) are used to georeference the images (Ai et al., 2015; Turner et al., 2014, 2012). “Geotags” are contained in an associated image metadata and specifies the UAVs position (coordinates) at each location that an image was taken. Upon loading the images, most processing software can automatically detect the camera used and its characteristics as well as automatically load the geotags. If GCPs were not measured, it is still possible to process the images into a seamless image using the geotags. As explained earlier, this will very likely lead to imprecise geolocation due to the normally low accuracies obtainable with consumer grade GPSs. The use of geotags only can cause your image to be shifted in a certain direction relative to accurately georeferenced data.
If GCPs were measured, they can be incorporated at this stage to improve the accuracy of the georeferencing. The minimum number of GCPs required is three, but these must not be on a line (horizontal or vertical) or clustered in one corner of the mission area. The more the GCPs, the better, although at a point adding more may not drastically improve accuracy estimates. Determining the optimal number of GCPs to use can be tricky. However, three factors can be considered in determining it: (1) size of the mission area, (2) magnitude of overlap and (3) level of visual content. In general, (1) the bigger the mission area, the higher the number of GCPs (evenly spread), (2) the larger the overlaps, the fewer the GCPs required, (3) the lower the visual content, the more GCPs are required. Visual content is affected by the availability of light, e.g. images taken in the dark or evening would have a low visual content. The reader is advised to consult the software manual of the processing software being used.
When incorporating GCPs at the geoferencing stage, care must be taken not to specify the wrong coordinate system in which the coordinates were measured or interchange/invert the “X” and “Y” coordinates.
Apart from proprietary software such as AgiSoft and Pix4D, there are open source applications that can be used for stitching UAV images. Examples are Microsoft Image Composite Editor (ICE), Visual SfM and OpenDroneMap. Microsoft ICE, for example, was found to be a useful and relatively easy-to-use application for image stitching although it is unable to produce digital surface model (DSM) and subsequently orthophotos. One may, for instance, choose to use it for stitching and later exporting the results to other processing applications for subsequent processing.
The International Potato Center (CIP) is developing an open source application for image stitching (ISAM-CIP V2). This software is focused on agricultural applications. Readers can obtain a trial copy here. Present limitations of this application are:
- limited range of supported output formats (currently .bmp) and
- no support for geographical information integration.
On the other hand, the use of the Pix4D software comes with these challenges when stitching the images from the multiSPEC4C. When low overlap images (less than 60% - altitudes of flight lower than recommended) are processed, it can happen that one of the four bands doesn’t stitch properly or it presents artifacts due to the image mismatch. A number of solutions can be pursued to minimize this problem. These are:
- to add manual tie points between the images that show inconsistencies,
- add GCPs that include latitude, longitude and height,
- or planning of the flights with 80% overlap or more. When high overlap can’t be achieved in one flight, it is possible to generate a perpendicular flight over the same area of interest that generates more images for a better stitching. Figures 5.8 and 5.9 are snapshots of aligned UAV images in AgiSoft and Pix4D Mapper software respectively.