Imaged based species recognition system

Perera, Thushara; Collins, John J.

doi:10.1109/icsenst.2015.7438391

Cited by 6 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This 79 allows us to document the spatial distribution of species-typical information across guenon faces 80 by identifying which face regions, when obscured, cause the break-down of correct species 81 classification. Eigenface decomposition was originally developed for individual face 82 discrimination in humans [32]; feature detection based on eigenfaces is also applicable to other 83 types of discrimination tasks involving complex animal signals [33][34][35] and has been used 84 previously to quantify guenon facial variation [10]. The perceptual face space generated by 85 eigenface decomposition parallels mammalian visual processing [36], lending biological 86 credibility.…”

Section: Introduction 26mentioning

confidence: 99%

The structure of species discrimination signals across a primate radiation

Winters

Allen

Higham

2019

Preprint

View full text Add to dashboard Cite

8Discriminating between conspecifics and heterospecifics potentially challenging for closely 9 related sympatric species. The guenons, a recent primate radiation, exhibit high degrees of 10 sympatry and form multi-species groups in which hybridization is possible but rare in most 11 populations. Guenons have species-specific colorful face patterns hypothesized to function in 12 species discrimination. Here, we apply a novel machine learning approach to identify the face 13 regions most essential for correct species classification across fifteen guenon species. We then 14 demonstrate the validity of these computational results using experiments with live guenons, 15showing that facial traits identified as critical for accurate classification do indeed influence 16 selective attention toward con-and heterospecific faces. Our results suggest variability among 17 guenon species in reliance on single-trait-based versus holistic facial characteristics when 18 discriminating between species, and differences in behavioral responses to faces can be linked to 19 whether discrimination is based on a single trait or whole-face pattern. Our study supports the 20 hypothesis that guenon face patterns function to promote species discrimination and provides 21 novel insights into the relationship between species interactions and phenotypic diversity. 22Keywords: species discrimination, visual signal form, complex signals, reproductive isolation, 23Cercopithecini 24 25 forage together [20]. Many guenons therefore interact with heterospecifics that share general 46 patterns of morphology (e.g. overall body size/shape) and behavior (e.g. activity patterns). In 47

show abstract

Section: Introduction 26mentioning

confidence: 99%

The structure of species discrimination signals across a primate radiation

Winters

Allen

Higham

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…In this research, we are investigating for implementation of a possible solution; which can be used to identify species. The eigenface technique is a novel approach in animal recognition domain [1,2]. In this application, cats and possums are targeted as species of interest.…”

mentioning

confidence: 99%

Image based Cats and Possums Identification for Intelligent Trapping Systems

Perera¹,

Collins²

2017

IJCA

Self Cite

View full text Add to dashboard Cite

Eigenface based recognition is widely used to identify individual faces. This technique is commonly used in human recognition domain. In this research, we are investigating for implementation of a possible solution; which can be used to identify species. The eigenface technique is a novel approach in animal recognition domain [1,2]. In this application, cats and possums are targeted as species of interest. The main reason for targeting cats and possum is, these introduced species are classified as threat to New Zealand natural environment and native wildlife. The next phase of this research project is to implement a vision based trapping system. This paper explains a novel classifier to use with eigenface techniques to separate different species.

show abstract