Anomaly Detection in EEG Signals: A Case Study on Similarity Measure

Figure 5

Results of six metrics using the training data of our database. From left to right: the similarity between each piece of data in the training data set and the template set and the accuracy of the metric for the normal training data, abnormal training data, and all training data. Detection result by using (a) ED, (b) PCCD, (c) SKLD, (d) HD, (e) KD, and (f) BD.

(a)

(b)

(c)

(d)

(e)

(f)