Method and System for Predicting of Road Surface Condition based on Machine Learning

摘要

The present invention relates to a machine learning-based road surface condition prediction method and system, and divides a picture taken in a predetermined area into a plurality of segments, the plurality of segments corresponding to a plurality of sub-regions belonging to the predetermined area, respectively; Spatial clustering is performed on a plurality of segments, and as a result of spatial clustering, the same cluster index value is assigned to a sub-region corresponding to a segment grouped into the same cluster, and a plurality of sub-regions are applied to the road surface condition data and weather data of the plurality of sub-regions. A step of generating training data mapped to a cluster index value assigned to a region for each of a plurality of segment size levels, and learning a plurality of road surface condition prediction models using the training data generated for each of the plurality of segment size levels, respectively including the steps of According to the present invention, by processing weather information and road surface condition information into characteristics including spatio-temporal information and learning a machine learning model, the accuracy of predicting road surface conditions for non-observed areas can be increased.