The satellite remote Sea Surface Height(SSH) images from 1992 to 2008 were analyzed, and the difficulty of detecting mesoscale eddies automatically was summarized. To improve the detection probability, a extracting mesoscale eddy method from the SSH images was proposed for the China seas and their adjacent seas. Combined with the characteristics of satellite remote SSH images, the background statistical characteristics were estimated by Gauss fitting method. According to the Gauss fitting results, the sea clutter background area was eliminated, thus the sea height abnormal area and a few of remainder sea clutter background area were derived. Then, according to the imaging characteristics of mesoscale eddies, a sets of detect algorithms for mesoscale eddies were designed combined with the criteria of intension and scale to distill mesoscale eddies in China seas and their adjacent seas adaptively. Finally, The detected mesoscale eddies were compared to those commented in literatures, and results prove that the algorithms have high detecting quality. The proposed method provides basis for analyzing space-time characteristics for mesoscale eddies in China seas and their adjacent seas across-the-board.