The phenomenon of micro-cracks on the sheds is due to aging of the composite insulator. With the development of the Unmanned Aerial Vehicle (UAV) technology, the online aging characterization based on micro-cracks turns out to be a reality. In this article, the micro-cracks identification and characterization on the sheds of composite insulators based on visible spectrum image was studied. At first, the shed area was segmented after graying and binarization operation. Then, the Laplace of Gaussian (LoG) operator and Otsu algorithm were used to extract the texture of micro-cracks. Box-counting method was used to calculate the fractal dimension of the micro-cracks on the composite insulator shed. The index of fractal dimension, a number between 1 and 2, can be used to describe the roughness and density of fractal dimension. The higher the fractal dimension was, the more severe the micro-cracks would be. The experimental results proved that the method proposed was feasible and effective on micro-cracks identification and characterization.