Measuring and tracking eye movements of a behaving archer fish by real-time stereo vision
| Autor: | Ohad Ben-Shahar, Avi Ben-Simon, Ronen Segev |
|---|---|
| Rok vydání: | 2009 |
| Předmět: |
Optics and Photonics
Time Factors genetic structures Eye Movements Rotation Computer science Movement Toxotes chatareus Motion Perception Video Recording Visual Acuity Neurophysiology Image processing Fixation Ocular Eye Species Specificity Image Processing Computer-Assisted Animals Computer vision Swimming Vision Binocular biology Behavior Animal business.industry General Neuroscience Fishes Eye movement Ethology biology.organism_classification Gaze Target acquisition Perciformes Stereopsis Motor Skills Predatory Behavior Space Perception Fixation (visual) Visual Perception Eye tracking Artificial intelligence business Artifacts Algorithms |
| Zdroj: | Journal of neuroscience methods. 184(2) |
| ISSN: | 1872-678X |
| Popis: | The archer fish (Toxotes chatareus) exhibits unique visual behavior in that it is able to aim at and shoot down with a squirt of water insects resting on the foliage above water level and then feed on them. This extreme behavior requires excellent visual acuity, learning, and tight synchronization between the visual system and body motion. This behavior also raises many important questions, such as the fish's ability to compensate for air-water refraction and the neural mechanisms underlying target acquisition. While many such questions remain open, significant insights towards solving them can be obtained by tracking the eye and body movements of freely behaving fish. Unfortunately, existing tracking methods suffer from either a high level of invasiveness or low resolution. Here, we present a video-based eye tracking method for accurately and remotely measuring the eye and body movements of a freely moving behaving fish. Based on a stereo vision system and a unique triangulation method that corrects for air-glass-water refraction, we are able to measure a full three-dimensional pose of the fish eye and body with high temporal and spatial resolution. Our method, being generic, can be applied to studying the behavior of marine animals in general. We demonstrate how data collected by our method may be used to show that the hunting behavior of the archer fish is composed of surfacing concomitant with rotating the body around the direction of the fish's fixed gaze towards the target, until the snout reaches in the correct shooting position at water level. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect