Method and Apparatus of Deep Learning-based Image Rotation and Motion-blur Correction

摘要

The present invention can be applied to an image recognition system, and an improvement in the image recognition recognition rate can be induced through the connected structure of the deep convolutional neural network algorithm and the denoising autoencoder algorithm. Through the present invention, it is possible to perform effective image recognition even in a situation where various movements of a subject and a camera recognition rate decrease.