A target re-identification method and system for a model organism tracking system

Abstract

The invention relates to a target re-identification method and system of a model organism tracking system. The method includes: acquiring multiple frames of continuous images under a first experimental condition, extracting a target connected domain in each frame of the image; extracting each isolated connected domain in the target connected domain, and extracting the target according to each isolated connected domain extract the intersection area of ​​the isolated connected domains of two adjacent frames; determine the primary association target according to the intersection area; use the Hungarian algorithm to process the remaining targets to determine the secondary association target; according to the primary association target and the secondary association target , generate a continuous image sequence for the same single target in different frame images, and train the convolutional neural network; the first frame of image under the second experimental condition, and use the trained convolutional neural network, so that the first frame of image in the image The target corresponds to the same target in the first experimental condition. The present invention improves tracking efficiency while performing target recognition under different experimental conditions.

Description

technical field [0001] The invention relates to the field of model organism tracking systems, in particular to a target re-identification method and system for a model organism tracking system. Background technique [0002] Accurately tracking the movement of model organisms is the basis for the operation of the micromanipulation system, and it is also an important means to verify the effect of micromanipulation. In biology, poisoning experiments and stress response experiments are often performed. Drugs are placed in petri dishes with model organisms or external stimuli are applied to track the movement of model organisms and responses to stimuli, and the effect of drugs is evaluated by analyzing the trajectory. and study the physiological functions of model organisms. [0003] In the field of model organism tracking, a two-stage tracking algorithm is commonly used for a single video. The first step is to detect each target (such as obtaining the position of each target, h...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a target re-identification method and system for a model organism tracking system to solve the problem that when the number of targets is large, the algorithm operation efficiency drops sharply, and it can only achieve target tracking for a single video, but cannot track different videos. The problem of identifying the target, which makes it impossible to compare the behavior patterns of the same target under different experimental conditions

Images