METHOD AND SYSTEM OF LEARNING A MODEL THAT AUTOMATICALLY DETERMINES DAMAGE INFORMATION FOR EACH PART OF AN AUTOMOBILE BASED ON DEEP LEARNING

摘要

The present invention relate to a deep learning based model learning method for automatically determining the degree of damage for each vehicle part and a system thereof which generate a model which can quickly produce a consistent and reliable estimate of vehicle repair by learning to automatically extract photos that can determine the degree of damage from accident vehicle photos using a deep learning-based mask R-CNN framework and an inception V4 network structure and learning the degree of damage for each type of damage.