Research Article

Human Gait Analysis: A Sequential Framework of Lightweight Deep Learning and Improved Moth-Flame Optimization Algorithm

Table 2

Proposed classification results of human gait recognition on the TUM GAID dataset.

ClassifierClass-based accuracy (%)Mean accuracy (%)
Normal walkWalk with a bagWalk with shoes

LightweightDeep-ELM99.6498.5297.6598.60
LightweightDeep-SVM98.5896.9296.2597.25
LightweightDeep-FKNN98.6396.2195.3796.73
LightweightDeep-EBT98.1296.8295.8096.91
LightweightDeep-DT97.5296.0395.2496.26

Bold values indicate the best values.