Zebrafish Larva Heart Localization Using a Video Magnification Algorithm

The heart position is an important and challenging step in the biological assays using zebrafish larvae for drug development. This paper presents a visual detection system for zebrafish larval heart position detection. The system develops a video magnification algorithm to solve this challenging problem in heart position detection: the heart signal of zebrafish larvae in the microscopic field is weak. It is difficult, if not possible, to be directly identified. The video magnification algorithm is used to amplify the heart dynamic information of zebrafish larvae so that the heart signal can be directly recognized by the proposed foreground detection algorithm, and then the integrated multi-frame image information is processed to obtain the position coordinates of the zebrafish larva heart. Finally, the proposed algorithm are tested and validated by zebrafish larval heart position detection experiments using the in-house developed micro-robotic system.