Design of English video course platform based on machine learning and Sobel algorithm

English video teaching is an indispensable part of English teaching, but there are certain difficulties in the design of English video courses, which need to be guided by artificial intelligence methods. Based on machine learning algorithm and Sobel algorithm, this paper builds an English video classroom design platform, uses image technology to process video, and proposes an improved algorithm based on Sobel operator. Moreover, this paper uses Sobel horizontal and vertical detection operators to define the edge strength of the block to be repaired and introduces a weight factor to enable the improved algorithm to repair images with different texture levels. In addition, this article improves the original SSD by introducing STD based on the image standard deviation and uses the degree of gray value dispersion between image blocks as a criterion for comparing whether two image blocks are similar, which effectively improves the accuracy of the search strategy. The experimental results show that the algorithm constructed in this paper can play a certain effect in teaching.