APPARATUS FOR TRAINING ARTIFICIAL NEURAL NETWORK DETECTING TARGETS IN HYPERSPECTRAL IMAGE AND METHOD THEREOF, APPARATUS FOR PROCESSING HYPERSPECTRAL IMAGE USING ARTIFICIAL NEURAL NETWORK

摘要

The learning device of an artificial neural network that detects a target in a hyperspectral image is a synthetic reflectance generator that generates a synthetic reflectance spectrum by linearly combining the eigenvectors of the reflectance spectrum stored in the reflectance spectral library, and the target reflectance spectral library. A reflectance spectral library for learning to generate learning data including a synthesized reflectance spectrum corresponding to the target reflectance spectrum by collecting a target reflectance spectrum and the synthesized reflectance spectrum, and calculating a radiation luminance from the learning data, and calculating the radiation luminance as a first And a first forward modeling unit for learning the artificial neural network by converting the apparent reflectance and the second apparent reflectance.