I³S is an acronym of Interactive Individual Identification System. I³S Manta is an extension of the system using the spot shape and size of the animals and is especially useful for identifying manta rays. This title explains most of I³S Manta’s functionality. First, we will discuss the interactive part. I³S Manta requires user interaction and is meant to support but not to replace the researcher. Initially, the user has to point out the most distinguishing features (i.e. the spots and certain reference points) of the unknown individual animal. In the next step, I³S Manta assists the user in the tedious task of matching animals for identification purposes. It automatically matches an annotated image with all other annotated images in the database and shows a ranked list of images. However, the user will always be responsible for making the final match between the unknown image and an image from the database.
The original version of I³S was designed for identification of ragged tooth sharks. The recognition procedure is based on the center location of the spots on the shark’s sides, and its dorsal and pelvic fins. I³S also proved to be useful for other species such as whale sharks. However, I³S turned out to be less suitable for species with very irregular spots, e.g. manta rays. To accommodate recognition of these types of species I³S was adapted. The new program was named I³S Manta. Apart from the spot location, I³S Manta also incorporates spot shape and size. Where I³S expected the user to point out the centers of the spots, I3S Manta requires from the user to draw an ellipse around each spot. Therefore, in I³S Manta the user interaction has slightly increased. However, because of the extra information about spot shape and size, the precision of the recognition algorithm has increased considerably as well. I³S should be used for species with regular spots. In case of considerable spot variation I³S Manta is advised.

The identification procedure assumes that the spot pattern of each individual animal is a unique distinguishing feature. The user points out the most distinguishing spots of each image. This spot pattern is stored in what is called a ‘fingerprint file’. You recognize these files by the extension .fgp. In the identification step, the resulting spot pattern is matched with the spot patterns of all the known animals in the database. To be able to correct for differences in viewing angle and scaling, the user is always required to point out three fixed reference points. With the reference points selected it is now possible to correct for rotation and scaling and compare two images in roughly the same 2D space or coordinate system. A comparison of two images comes down to finding corresponding spot pairs in the above-mentioned coordinate system. Further, the spot sizes and the ratio between length and width of both ellipses should be sufficiently similar.