Computational Intelligence and Neuroscience / 2021 / Article / Alg 4

Research Article

An Efficient Automatic Gait Anomaly Detection Method Based on Semisupervised Clustering

Algorithm 4

The pseudocodes of the gait classification.
Input: (1) Disjoint λ sets of gait feature vectors with preassigned class labels
      (2) Gait feature vector set of the individuals involved in the previous run
      (3) The gait features vector of the current detecting individuals
Output: the class labels of .
1. Cluster the with BC-COP-K-means algorithm with K = 2
2. If is assigned to the cluster with fewer members
  Classify to the group of individuals with abnormal gaits.
  Classify to the group of individuals with normal gaits.