Method and system for detecting light spot track of single-fiber scanner based on deep learning

Abstract

The invention provides a method and system for detecting a light spot track of a single-fiber scanner based on deep learning, and the method comprises the steps: collecting original data, making a data set, training a model, and using the model, and inputting the original data collected in real time into the model trained by a network to obtain light spot scanning track prediction data. Correspondingly, the invention further provides a system for detecting the light spot track of the single-fiber scanner based on deep learning so as to realize the method. According to the invention, real-time detection of the light spot track of the single-fiber scanner is realized, the problem of distortion of an image shot or projected by the single-fiber scanner is solved, and the precision and the anti-interference capability of the system are improved.

Description

technical field

[0001] The present invention relates to the field of optical scanning technology, in particular to a method and system for detecting a spot track of a single optical fiber scanner based on deep learning, which can be used in the fields of medical imaging and photoelectric detection. Background technique

[0002] A single fiber scanner is a device that drives a single fiber to swing through an actuator, so that the light spot emitted from the fiber cantilever scans a certain range on the target. It can be used for endoscopes, micro-projection and three-dimensional scanning detection, etc. field. Among them, how to obtain the spot motion track in real time and accurately is the difficulty and focus in the single-fiber scanning technology. At present, the mature control scheme in the single-fiber scanning technology adopts the research report (High Performance Open Loop Control of Scanning with a Small Cylindrical Cantilever Beam , Journal of Sound and Vibratio...

Images