Research Article

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

Table 1

Proposed classification results of human gait recognition on the CASIA B dataset.

MethodClass18°36°54°72°90°108°126°144°162°180°Mean

LightweightDeep-ELMNM97.198.29593.898.197.598.397.2989498.696.89
BG94.295.39192.79389.794.893.192.891.895.493.07
CL78.883.582.186.378.590.285.98281.388.383.483.66

LightweightDeep-SVMNM96.297.596.192.897.996.29897.597.193.29896.40
BG92.593.69392.294.187.69493.592.192.493.492.58
CL7982.881.58679.187.483.982.380.787.582.182.93

LightweightDeep-FKNNNM93.294.592.993.194.691.893.594.894.59394.193.63
BG8789.190.589.691.482.690.889.48789.390.788.85
CL73.578.477.281.675.382.28077.576.182.478.578.42

LightweightDeep-EBTNM92.893.390.491.59592.692.493.19493.792.992.88
BG88.587.490.190.591.88290.290.486.189.891.988.97
CL72.680.17780.47581.680.676.278.58278.178.37

LightweightDeep-DTNM87.488.990.191.69390.58891.291.387.490.589.99
BG81.582.687.583.890.480.69087.985.38490.185.79
CL69.872.17778.572.78078.372.572.980.180.275.82

Bold values indicate the best values.