摘要
本發明提出一種資料增強方法，用於擴展包括多個時頻譜（Spectrogram）的資料集。上述方法包括在資料集中的第一時頻譜中選擇至少一個區塊；決定所選的至少一個區塊的至少一個調整值；以及基於至少一個調整值調整第一時頻譜的至少一個區塊，以得到第一調整後時頻譜。上述調整值包括對比度調整值、亮度調整值以及伽瑪調整值的至少其中之一。
A data augmentation method for expanding a dataset including a plurality of spectrograms is provided. The method includes selecting at least one patch of a first spectrogram of the plurality of spectrograms, determining at least one adjustment value corresponding to the at least one patch, and adjusting the at least one patch of the first spectrogram based on the at least one adjustment value, to obtain a first adjusted spectrogram. The at least one adjustment value includes at least one of a contrast adjustment value, a brightness adjustment value, and a gamma adjustment value.