The feature extraction method of students' ideological and political thinking online learning behavior based on visual cognition

Abstract

In order to solve the problem of inaccurate extraction results of students' ideological and political online learning behavior feature points in traditional online learning behavior feature extraction methods, a small number of correct recognitions, and a long extraction time, a visual cognition-based student ideological and political online learning behavior feature is proposed Extraction Method. Analyze the basic operating principles of visual cognitive technology, and build a corresponding visual cognitive computing model. Under this model, the students' remote ideological and political teaching system is used as a platform to collect real-time behavior data, and the standardization of initial image data is realized from two aspects: grayscale and geometry. In the processed data, the visual cognitive computing model is used to determine the target position of the student's learning behavior in the screen, and the target moving contour and blinking behavior characteristics are extracted within the location area, so as to realize the extraction of the behavior characteristics of students' ideological and political online learning. Simulation experiments show that the method has a large number of correct recognition of students' ideological and political online learning behavior characteristics, the extraction results are more accurate, and the extraction efficiency is high.