Earthquakes can be detected by matching spatial patterns or phase properties from 1D seismic waves. Current earthquake detection methods such as waveform correlation and template matching (TM) have difficulty detecting anomalous earthquakes that are not similar to other earthquakes. In recent years, machine‐learning techniques for earthquake detection have been emerging as a new active research direction. In this article, we develop a novel earthquake detection method based on dictionary learning. Our detection method first generates rich features via signal processing and statistical methods, and further employs feature selection techniques to choose features that carry the most significant information. Based on these selected features, we build a dictionary for classifying earthquake events from nonearthquake events. To evaluate the performance of our dictionary‐based detection methods, we test our method on a labquake dataset, which contains 3,357,566 time‐series data points with a 400 MHz sampling rate. A total of 1000 earthquake events are manually labeled, and the length of these earthquake events varies from 74 to 7151 data points. Through comparison with other detection methods, we show that our feature selection and dictionary‐learning incorporated earthquake detection method achieves an 80.1% prediction accuracy and outperforms the baseline methods in earthquake detection, including TM and support vector machine (SVM).