A Cell Phone-based Sign Reader for Blind & Visually Impaired Persons

描述（由申请人提供）：我们建议开发和评估一种基于手机的系统，使盲人和视障人士能够找到和阅读街道标志和其他与寻路有关的标志。利用标准手机的内置摄像头和计算能力，该系统将处理用户捕获的图像，以查找和分析标志，并说出其内容。这将为盲人或视障行人在寻找和阅读街道标志，以及定位和识别地址和商店名称方面提供宝贵的帮助，而无需他们携带任何专用硬件。这款标识识别和读取软件将免费提供给任何使用广泛使用的塞班操作系统（如流行的诺基亚手机系列）的配备摄像头的手机。我们将以我们之前和正在进行的工作为基础，将计算机视觉技术应用于盲人的实际问题解决，包括室内寻路算法的手机实现和数字设备显示的阅读。我们将开发、完善并将一种新的基于信念传播的算法转移到手机平台上，该算法已在包括部分阴影覆盖在内的困难现实条件下发现和分析信号方面取得了初步成功。人为因素研究将有助于确定如何配置系统及其用户控制以获得最大的有效性和易用性，并提供对整个系统的评估。对于视力正常的人来说，每天都能看到环境标志、标志或地标是理所当然的，但在这项任务中，大约有1000万严重视力障碍的美国人和100万法定失明的美国人面临着严重的困难。拟议的研究将产生一个高度可访问的系统（用户的成本为零或最低），以增强现有的寻路技术，这将极大地改善盲人和视障人士的独立旅行。

DESCRIPTION (provided by applicant): We propose to develop and evaluate a cell-phone-based system to enable blind and visually impaired individuals to find and read street signs and other signs relevant to wayfinding. Using the built-in camera and computing power of a standard cell phone, the system will process images captured by the user to find and analyze signs, and speak their contents. This will provide valuable assistance for blind or visually impaired pedestrians in finding and reading street signs, as well as locating and identifying addresses and store names, without requiring them to carry any special-purpose hardware. The sign finding and reading software will be made freely available for download into any camera-equipped cell phone that uses the widespread Symbian operating system (such as the popular Nokia cell phone series). We will build on our prior and ongoing work in applying computer vision techniques to practical problem-solving for blind persons, including cell-phone implementation of algorithms for indoor wayfinding and for reading digital appliance displays. We will develop, refine and transfer to the cell phone platform a new belief propagation-based algorithm that has shown preliminary success in finding and analyzing signs under difficult real-world conditions including partial shadow coverage. Human factors studies will help determine how to configure the system and its user controls for maximum effectiveness and ease of use, and provide an evaluation of the overall system. Access to environmental labels, signs or landmarks is taken for granted every day by the sighted, but approximately 10 million Americans with significant vision impairments and a million who are legally blind face severe difficulties in this task. The proposed research would result in a highly accessible system (with zero or minimal cost to users) to augment existing wayfinding techniques, which could dramatically improve independent travel for blind and visually impaired persons.