Upload Beetle Images

The frontal aspect can be informative, although variable positioning of the antennae and mouthparts means that this aspect will improve as more data are added to train the AI model.

Currently the model performs best if the image is captured with the frons (the area between the eyes) in alignment with the clypeus.

The dorsal habitus has been shown to have significant information in both geometric morphometrics (through the shape of both the pronotum and the elytra) and in informing the training of the AI model for identification.

We are training the model with variation in the level of the head with the pygidium, however the images should be uploaded with the head at the left.

The caudal habitus has provided significant information for accurate identification in testing and training the model.

While we have attempted to train the model with some variation of positioning the specimen, it will work best if the image is captured with the pygidium (the last dorsal plate of the abdomen) viewed in a single plane.

The lateral aspect can be informative, although variable positioning of appendages means that this aspect will improve as more data are added to train the AI model.

We are training the model with variation in the level of the head with the pygidium, however the images should be uploaded with the head at the left.