Real-time biological image recognition method and device

Abstract

The invention discloses a biological image recognition device and method. The biological image recognition device comprises a processor and a memory. The memory comprises at least one instruction which is executed by the processor; wherein the processor generates fusion data using any one biological image at any point in time among biological images continuously captured in time with respect to a subject, and sensor data temporally corresponding to the continuously captured biological images; and extracting attribute information of the continuously captured biological images from the fused data based on a machine learning model. Therefore, the biological image recognition device can recognize the object body in the real-time image based on the machine learning model independent of the performance of the memory or the processor. The invention relates to a technology developed through the technical commercialization support project (TB210264) of the first industry revitalizing yard of the first city in 2021 'Instructions for detecting the functions of explaining artificial intelligence products of main organs in intraperitoneal endoscopic surgery'.

Description

technical field [0001] The present invention relates to methods of identifying real-time images, and more particularly, to a method and apparatus capable of tracking tissue and predicting tissue location within real-time biological images. Background technique [0002] Recently, with the development of artificial intelligence learning models, a large number of machine learning models are being utilized in interpreting images. For example, learning models such as Convolutional Neural Networks, Deep Neural Networks, Recurrent Neural Networks, Deep Belief Networks are being applied to Still Image or real-time Image (Motion Picture) detection, classification and feature learning. [0003] When using machine learning models to learn real-time images (videos), the amount of learning and computation on image data is huge. If the processor performance or memory capacity is not high enough, it is difficult to identify objects in real-time images. SUMMARY OF THE INVENTION [0004]...

AI Technical Summary

Problems solved by technology

[0003] When using a machine learning model to learn real-time images (videos), the amount of learning and calculation of image data is huge. If the processor performance or memory capacity is not high enough, it is difficult to recognize objects in real-time images.

Images