Abstract:Brood parasitized and/or colonial birds use egg features as visual identity signals, which allow parents to recognize their own eggs and avoid paying fitness costs of misdirecting their care to others' offspring. However, the mechanisms of egg recognition and discrimination are poorly understood. Most studies have put their focus on individual abilities to carry out these behavioural tasks, while less attention has been paid to the egg and how its signals may evolve to enhance its identification. We used 92 clutches (460 eggs) of the Eurasian coot Fulica atra to test whether eggs could be correctly classified into their corresponding clutches based only on their external appearance. Using SpotEgg, we characterized the eggs in 27 variables of colour, spottiness, shape and size from calibrated digital images. Then, we used these variables in a supervised machine learning algorithm for multi-class egg classification, where each egg was classified to the best matched clutch out of 92 studied clutches. The best model with all 27 explanatory variables assigned correctly 53.3% (CI = 42.6-63.7%) of eggs of the test-set, greatly exceeding the probability to classify the eggs by chance (1/92, 1.1%). This finding supports the hypothesis that eggs have visual identity signals in their phenotypes. Simplified models with fewer explanatory variables (10 or 15) showed lesser classification ability than full models, suggesting that birds may use multiple traits for egg recognition. Therefore, egg phenotypes should be assessed in their full complexity, including colour, patterning, shape and size. Most important variables for classification were those with the highest intraclutch correlation, demonstrating that individual recognition traits are repeatable. Algorithm classification performance improved by each extra training egg added to the model. Thus, repetition of egg design within a clutch would reinforce signals and would help females to create an internal template for true recognition of their own eggs. In conclusion, our novel approach based on machine learning provided important insights on how signallers broadcast their specific signature cues to enhance their recognisability.

Identification of Pelagic Eggs by Image Analysis

Image Classification of Chicken Embryo Based on Matched Filter and Skeleton Curvature Feature

A Real-Time Oocyte Polar Body Detection Method

Lipid Metabolic Heterogeneity During Early Embryogenesis Revealed by Hyper-3D Stimulated Raman Imaging

K-Means Segmentation Based-on Lab Color Space for Embryo Detection in Incubated Egg

Egg recognition: The importance of quantifying multiple repeatable features as visual identity signals

Automatic stage identification of Drosophila egg chamber based on DAPI images

Eggshell Biometrics for Individual Egg Identification Based on Convolutional Neural Networks.

Sex Detection in the Early Stage of Fertilized Chicken Eggs via Image Recognition

An approach for goose egg recognition for robot picking based on deep learning

Molecular Identification of Species Diversity Using Pelagic Fish Eggs in Spring and Late Autumn-Winter in the Eastern Beibu Gulf, China

Egg Weight Detection on Machine Vision System

Lipid Droplet Composition Varies Based on Medaka Fish Eggs Development as Revealed by NIR-, MIR-, and Raman Imaging

Classification of Overlapping Eggs Based on Image Processing

Identification of eggs from different production systems based on hyperspectra and CS-SVM

Quantification of Intra Embryonic Motions Through Label Free and Fast Imaging Of Yolk Granules

Pattern Recognition of Eggshell Crack Using PCA and LDA

Monitoring storage time and quality attribute of egg based on electronic nose.

Prediction of Egg Storage Time and Yolk Index Based on Electronic Nose Combined with Chemometric Methods

Non-invasive laser speckle imaging of extra-embryonic blood vessels in intact few-days-old avian eggs

Automated contour extraction for light‐sheet microscopy images of zebrafish embryos based on object edge detection algorithm