CORIGAN: Assessing multiple species and interactions within images

Abstract

Images are resourceful data for ecologists and can provide a more complete information than other methods to study biodiversity and the interactions between species. Automated image analysis however often relies on extensive datasets, not implementable by small research teams. We are here proposing an object detection method that allows the analysis of high‐resolution images containing many animals interacting in a small dataset. We developed an image analysis pipeline named ‘CORIGAN’ to extract the characteristics of animal communities. CORIGAN is based on the YOLOv3 model as the core of object detection. To illustrate potential applications, we use images collected during a sentinel prey experiment. Our pipeline can be used to detect, count and study the physical interactions between various animals. On our example dataset, the model reaches 86.6% precision and 88.9% recall at the species level or even at the caste level for ants. The training set required fewer than 10 hr of labelling. Based on the pipeline output, it was possible to build the trophic and non‐trophic interactions network describing the studied community. CORIGAN relies on generic properties of the detected animals and can be used for a wide range of studies and supports. Here, we study invertebrates on high‐resolution images, but the same processing can be transferred for the study of larger animals on satellite or aircraft images.

The challenge of biological control of Cosmopolites sordidus Germar (Col. Curculionidae): A review

“…Poeydebat et al (2017) found a negative correlation between generalist predator abundance and C. sordidus abundance. In Costa Rica, cockroaches have been observed to feed on exposed C. sordidus larvae (Tresson, Tixier, Puech, Bagny Beilhe, et al, 2019). Vinatier et al (2009) also emphasized the effect of predators on the developmental stages of C. sordidus; the authors indicated that generalist predators and particularly egg predators help reduce banana weevil numbers.…”

“…Photographs taken in the field allowed detailed analysis of the behaviours of different ant species in Costa Rica (Tresson, Tixier, Puech, Bagny Beilhe, et al, 2019). Camponotus atriceps (Hymenoptera, Formicinae) were observed to feed on dead adults but not on any living stages of C. sordidus; S. geminata consumed C. sordidus eggs and larvae but were unable to kill a live adult; Pheidole radoszkowskii (Hymenoptera, Myrmicinae), O. bauri and Nylanderia sp.…”

“…consumed larvae and also were scavengers. Furthermore, a single individual of O. bauri was able to seize a larva, while at least five P. radoszkowskii were needed to capture a larva (Tresson, Tixier, Puech, Bagny Beilhe, et al, 2019). Ants show a great variety of behaviour, size and activities that are relevant to their role as biological control agents.…”

“…More recently, methods have been developed that rely on the combined use of sentinel prey and automated image analysis (Tresson, Tixier, Puech, Bagny Beilhe, et al, 2019;. Images as a source of information have the advantage of showing predation with certainty and with fewer behavioural biases than occur in laboratory experiments.…”

Caught on camera: Field imagery reveals the unexpected importance of vertebrates for biological control of the banana weevil (Cosmopolites sordidus Col. Curculionidae)

“…More recent uses include for example species detection and tracking [ 25 ] or interaction analysis [ 26 ]. The use of automated image analysis can allow a detailed study of the recruiting dynamics of ants or of the number of individuals needed to seize a prey animal, enabling the reconstruction of the observed interaction network [ 26 ]. In some cases, however, image analysis can be performed with less demanding methods, such as background subtraction or color thresholding, and yield sufficient results without the need of a specific labelled dataset.…”