IMAGE ENHANCEMENT METHOD USING NEURAL NETWORK MODEL BASED ON EDGE COMPONENT CLASSIFICATION

摘要

PURPOSE: An image enhancement method using a neural network model based on border line component classification is provided to efficiently restore a border line portion sensitive to the sight of a human. CONSTITUTION: An image for each frequency is generated in relation to an input image(S10), and border line images is classified according to each component in order to perform efficient learning depending on each component of an IF(Intermediate Frequency) image(S20). According to the classified border line image, a neural network model is configured(S30). Through the neural network model, a high frequency image is estimated. Through the sum of the estimated high frequency image and low frequency image, a high-definition image is implanted(S40).