The present invention relates to a method of acquiring gait data using an insole including a pressure sensor array, normalizing the acquired gait data, and automatically classifying a gait type using machine learning, included in a smart insole A walking data collection unit that acquires walking data from a plurality of pressure sensors, and a section of unit steps are defined in the walking data acquired by the walking data collection unit, so that the data for each unit step is divided, and the data is divided so that the lengths of all the data are the same. It includes a pre-processing unit to normalize (normalize), a walking pattern classification unit to classify the final walking type using a fully connected network (fully connected network) that receives the normalized walking data as input from the pre-processing unit.
展开▼