A computer-based method for intelligently identifying the state of human eyes in videos based on energy changes in gray-level co-occurrence matrices

Abstract

The invention discloses a method for intelligently identifying the human eye state in a video by adopting the energy change of a gray scale co-occurrence matrix, comprising the following steps: (1) acquiring several human eye images from the video, and calculating the gray scale of each human eye image Co-occurrence matrix; (2) calculate the angular second-order moment value of the gray-level co-occurrence matrix corresponding to each frame of human eye image, and obtain the corresponding energy value of each frame of human eye image; construct a raw data curve of frame number-energy value; (3) Perform mean value filtering on each frame of the human eye image to obtain a reference curve corresponding to the original data curve; subtract the corresponding values ​​on the original data curve and the reference curve and take the absolute value to obtain a difference curve ; (4) Compare the difference on the difference curve with the preset energy threshold. This method has strong adaptability and can effectively overcome the influence of differences in eyes of different people.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and in particular relates to a method for intelligently identifying the state of human eyes in a video by using the energy change of a gray scale co-occurrence matrix. Background technique [0002] Eyes are the most important feature of the human face, and play an extremely important role in the research and application of digital image processing and computer vision. The accuracy rate directly affects the performance of the system. However, in the actual application process, the influence of many factors such as uneven illumination, light spots, eyelashes and spectacle frames makes the recognition of eye opening and closing status a very challenging task. [0003] At present, there are many methods in the field of eye opening and closing state recognition, the most representative ones include sample learning method, template matching method, upper and lower eyelid detection m...

AI Technical Summary

Problems solved by technology

[0010] The present invention provides a method for intelligently identifying the state of human eyes in a video using the energy change of the gray-level co-occurrence matrix, which solves the problem of poor real-time performance and susceptibility to various factors commonly found in the prior art when the computer recognizes the state of eye opening and closing. Problems with impact and poor robustness

Images