Deep learning based classification system using image data augmentation, and cotrol method thereof

摘要

The present invention relates to a classification system using deep learning-based image data generation technology and a control method thereof, wherein the system according to the present invention compresses input image data and extracts features and input image data compressed by the encoder. It includes an image data generator including a decoder for generating reconstructed image data, and a convolutional neural network unit. A data classification unit for classifying input image data is constructed using an encoder and a convolutional neural network unit. According to the present invention, there is an advantage of overcoming the drawbacks when the input of deep learning is simply a defect image and maximizing the defect classification performance while minimizing the computation time.