A vision detection system, method and storage medium based on deep learning

Abstract

The invention discloses a vision detection system, method and storage medium based on deep learning. The vision detection system includes a logo display module, an image acquisition module, a posture evaluation module, and a result evaluation module. The invention displays the vision detection through the logo display module. To identify, the user uses the arm to make corresponding body movements; the image acquisition module is used to collect images of the body movements made by the user's arms, and input them to the posture evaluation module; the posture evaluation module is used to detect the body movements made by the collected arms Key points of posture: the result evaluation module judges the state of the user's arm according to the key points of the posture of the user's arm, and then judges whether the user's action is consistent with the vision detection mark, and outputs the detection result. The result evaluation module in the present invention judges whether it is consistent according to the direction of the user's arm posture. Compared with the direction of the finger, the target of the arm posture is larger and easier to identify, and the detection accuracy is higher.

Description

technical field [0001] The invention belongs to the technical field of vision detection, and in particular relates to a vision detection system, method and storage medium based on deep learning. Background technique [0002] With the advancement of science and technology and the popularity of information technology, people spend more and more time using high-tech equipment such as mobile phones, computers, and televisions, and the resulting risk of visual impairment has also increased, especially for young people. Can not be ignored. The usual visual acuity testing is generally carried out manually in professional institutions, and the user cannot perform independent operations. [0003] At the same time, some intelligent vision detection systems, such as the intelligent vision detection instrument based on image analysis disclosed in the patent CN106778597A, use the detector of the traditional algorithm in the field of machine vision to judge the posture and orientation of...

AI Technical Summary

Problems solved by technology

However, such methods are based on traditional algorithms with relatively low accuracy

At the same time, due to the small target and complex joints of the finger, it is particularly difficult to accurately judge the gesture and pointing of the finger, which limits the performance of the system.

[0004] In the field of deep learning, the machine vision model based on the self-attention mechanism that has emerged in recent years has achieved excellent accuracy in many fields. square exponential growth

Images