Auto-generating tactile from image for low vision and blind individuals

对于视障和盲人来说，获取印刷媒体一直是一项挑战。盲文书写系统和文本语音转换工具在使图片和图形可供这一人群使用方面存在根本限制。触觉已经被发现是交流图形信息的最有效的方式。触觉是一种表示，通过它可以通过触摸访问内容。该项目旨在开发一个基于机器学习的系统，自动从RGB图像生成触觉图像。该项目的产品将是一个软件，其输出是可以使用盲文压纹机打印的触觉图像。预计拟议的系统将有助于创建触觉的过程，特别是通过为触觉专家提供自动生成原型的工具来缩短设计时间。此外，该项目还将有助于提高机器学习研究的知识水平。

Accessing print media has been a challenge for visually impaired and blind people. Braille writing system and text-to-speech tools have fundamental limitations in making pictures and graphics accessible to this population. Tactile has been found to be the most effective way to communicate graphical information. Tactile is a representation through which content can be accessed by touch. This project aims to develop a machine learning-based system that automatically generates tactile images from RGB images. The product of this project will be a software whose output is a tactile image that can be printed using a braille embosser. The proposed system is expected to help with the process of creating tactiles, in particular by reducing the design time by providing tactile experts with tools for auto-generating prototypes. Besides, the project will contribute to the advancement of knowledge in machine learning research.