<p>The power line harmonic generated by human activities&#160;can be found from the vast amount of the data observed by EFD on board the ZH-1 satellite. To&#160;study&#160;the&#160;human activities and&#160;remove the nonnegligible amount of interferences in the study of ionospheric precursors of earthquakes, we are desperate&#160;for&#160;finding the power line harmonic from the vast amount of data Hence, a novel automatic&#160;power line recognition method is proposed.&#160;Firstly, we utilize fourier transform on EFD data to obtain the power spectral density(PSD).&#160;Secondly, it is well known that harmonic radiation from power lines presents one or more horizontal linear characteristics on the PSD image&#160;and the color of the line is close to the color of the background&#160;in the&#160;image.In order to highlight&#160;the&#160;color contrast between the line and the background, we transform the&#160;PSD&#160;image&#160;from the RGB to the HSV color space&#160;and utilize the&#160;Saturation&#160;compoment of the HSV space as the object image.To obtain the edge regions, we process the object image with canny techniques. Finally, we use the Hough transform&#160;to detect the power line from the edge regions. To evaluate the proposed method, the experiment is performed for the dataset composed of 100 PSD images&#160;and&#160;each PSD image includes several&#160;interference lines. And the experimental result verifies the effectiveness of the proposed method with an accuracy of 86%.</p>