Wind turbine clearance refers to the minimum distance that must be maintained between the rotating blades of the wind turbine and the tower where the blades are installed. This clearance is very important to ensure the normal operation of the wind turbine. If the distance is too small, it may cause tower sweeping (blade impact on the tower), so it is necessary to monitor the clearance to ensure the normal operation of the wind turbine. At present, many methods based on image processing have been proposed to calculate the wind turbine clearance. Since the position of the wind generator tower in the image hardly change, the difficulty of the wind turbine clearance calculation based on image processing lies in accurately giving the profile of the blade. In this context, this paper not only integrates the traditional algorithm and deep learning algorithm, but also proposes an expansion filtering algorithm to improve the detection effect of the blade. We evaluate our method on the collected data set, the result shows that the method proposed in this paper is better than the traditional algorithm and deep learning.