In a method for optimizing the image processing of web videos, information of an image frame in a web video is obtained. A canvas element based on the information of the image is created, and a canvas element is drawn based on the information of the image. Attached recognition information is obtained. A block element based on the attached information is created, and the attached information is set into the block element. The canvas element and the block element are composed and stacked on a web page. The canvas element and the block element are displayed in the web page. An electronic device, and a non-transitory computer readable storage medium applying the method are also provided.