Eye-Interface System Using Convolutional Neural Networks for People with Physical Disabilities
Eye-Interface System Using Convolutional Neural Networks for People with Physical Disabilities
复制标题
使用卷积神经网络为身体残疾人士提供眼部接口系统
DOI：
10.1007/978-3-030-70451-3_7
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Shenglin Mu
中科院分区：
文献类型：
--
作者：
Keiichiro Kubo;Satoru Shibata;Tomonori Karita;Tomonori Yamamoto;Shenglin Mu
近年来，缺乏有效的沟通方式是肢体残疾人（如肌萎缩侧索硬化症）在社会生活中面临的重要问题之一。在这项研究中，我们构建了一个基于图像处理方法的眼睛注视输入系统，使用PC和网络摄像头，价格低廉，易于安装在自然光环境下，没有疾病的风险，由于使用红外辐射。在我们提出的系统中，卷积神经网络（CNN）被应用于提高实际使用的准确性。本研究中的CNN根据从网络摄像机获取的图像估计监视器屏幕上的注视位置，并旨在通过针对特定个体的学习来获得比传统系统更高的精度。
In recent years, lack of effective communication method is one of the important issues that the people with physical disabilities, such as ALS (amyotrophic lateral sclerosis) face in social life. In this research we constructed an eye-gaze input system based on an image processing method using a PC and a web camera that are inexpensive and easy to install under natural light environment without the risk of illness due to the use of infrared radiation. In our proposed system, convolutional neural networks (CNN) is applied to improve the accuracy for practical use. The CNN in this study estimates the gaze position on the monitor screen from the image acquired from the web camera, and aims to obtain higher accuracy than the conventional system by learning for specific individuals.
通过眼球运动操作的肢体残疾人数据输入装置。
DOI：
10.1299/kikaic.63.1546
发表时间：
1997
期刊：
Transactions of the Japan Society of Mechanical Engineers. C
影响因子：
--
作者：
T. Ochiai;T. Ishimatsu;O. Takami;R. Matsui
通讯作者：
R. Matsui