Apparatus and method for image analysis using virtual three-dimensional deep neural network

Abstract

An apparatus for image analysis includes: an image acquisition unit for stacking a plurality of two-dimensional image data in a predetermined order; a three-dimensional image generation unit for generating a plurality of three-dimensional data on the basis of different types of multiple items of information for the plurality of two-dimensional image data in a stacked form from the image acquisition unit; and a deep learning algorithm analysis unit for applying a two-dimensional convolutional neural network to each of the plurality of three-dimensional data from the three-dimensional image generation unit, and combining results of applying the two-dimensional convolutional neural network to the plurality of three-dimensional data.

Description

TECHNICAL FIELD[0001]The present invention relates to an image analysis technique using image reconstruction, and more specifically, to an apparatus and method for image analysis is using a virtual three-dimensional deep neural network.BACKGROUND ART[0002]An artificial neural network (ANN) is one of techniques for implementing machine learning.[0003]Generally, the artificial neural network is configured of an input layer, a hidden layer, and an output layer. Each of the layers is formed of neurons, and neurons of each layer are connected to the output of the neurons of a previous layer. A value obtained by adding a bias to a value calculated through an inner product of the output values of the neurons of the previous layer and connection weights corresponding thereto is put into an activation function, which is generally nonlinear, and an output value of the activation function is transferred to the neurons of the next layer.[0004]Existing machine learning techniques learn a classif...

AI Technical Summary

Benefits of technology

[0022]According to the present invention, there is an advantage of more effectively learning three-dimensional data and analyzing images using a two-dimensional convolutional neural network having a small number of parameters compared with a general three-dimensional convolutional neural network method.

[0023]In addition, according to the present invention, it is possible to provide a new image analysis model which can perform effective learning and image analysis on three-dimensional image data, while solving the problem of a three-dimensional convolutional neural network model occupying a lot of memory since the number of parameters is large, taking a long time in learning, and having a long calculation time when using a learned model.

Problems solved by technology

Convolutional neural networks (CNNs) draw much attention, overwhelming the performance of existing machine learning techniques in the field of image recognition.

However, this technique provides a framework applying a basic deep learning model, which is somewhat different from constructing a model of a particular structure.

Images