Assessing the potential information content of multicomponent visual signals: a machine learning approach

Abstract

Careful investigation of the form of animal signals can offer novel insights into their function. Here, we deconstruct the face patterns of a tribe of primates, the guenons (Cercopithecini), and examine the information that is potentially available in the perceptual dimensions of their multicomponent displays. Using standardized colour-calibrated images of guenon faces, we measure variation in appearance both within and between species. Overall face pattern was quantified using the computer vision 'eigenface' technique, and eyebrow and nose-spot focal traits were described using computational image segmentation and shape analysis. Discriminant function analyses established whether these perceptual dimensions could be used to reliably classify species identity, individual identity, age and sex, and, if so, identify the dimensions that carry this information. Across the 12 species studied, we found that both overall face pattern and focal trait differences could be used to categorize species and individuals reliably, whereas correct classification of age category and sex was not possible. This pattern makes sense, as guenons often form mixed-species groups in which familiar conspecifics develop complex differentiated social relationships but where the presence of heterospecifics creates hybridization risk. Our approach should be broadly applicable to the investigation of visual signal function across the animal kingdom.

Heritable variation in colour patterns mediating individual recognition

“…Individuality is the result of combinatorial variation in multiple phenotypes in P. fuscatus wasps [3] and other taxa [13,49–52], which has led to the prediction that traits involved in individual identity should be genetically uncorrelated [16,17]. Here we find a mix of traits with mostly moderate genetic correlations (e.g.…”

Automatically detecting and tracking free‐ranging Japanese macaques in video recordings with deep learning and particle filters

“…Algorithms in previous studies have been used to detect or track facial regions in social animals such as primates and lions living in their natural habitats (Allen & Higham, ; Burghardt & Calic, ; Loos & Ernst, ). Although facial identification is a useful feature, insofar as the faces of primates can be used to derive various information (e.g., the sex and age of the animal) (Allen & Higham, ), tracking is only possible when the facial region is detected. In contrast to the previous studies, the algorithm in our model can detect body regions regardless of the presence of facial regions.…”

“…The technology for automated observation with AI includes animal detection and tracking from pictures or video recordings. Machine learning, a subfield of AI, is used to detect particular species or individuals, such as guenons (Allen & Higham, ), macaques (Witham, ), African great apes (Ernst & Kublbeck, ), zebras (Lahiri, Tantipathananandh, Warungu, Rubenstein, & Berger‐Wolf, ), and penguins (Burghardt, Thomas, Barham, & Calic, ). A machine learning system classifies the content of an image from raw data into groups that share similar properties.…”

patternize: An R package for quantifying colour pattern variation

“…Image segmentation concerns the categorization of pixels by colour. Previously, examples of colour pattern quantification have been extensively developed for Heliconius butterflies (Color Pattern Modelling [CPM] in Le Poul et al., ) and primates (Allen, Higham, & Allen, ). However, these applications are currently not easily accessible for use in other organisms.…”

“…However, precisely quantifying colour pattern variation is challenging. Consistent comparisons of colour patterns from images requires the (1) homologous alignment and (2) Modelling [CPM] in Le Poul et al, 2014) and primates (Allen, Higham, & Allen, 2015). However, these applications are currently not easily accessible for use in other organisms.…”