Mid-Level Vision Systems for Low Vision

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Low vision is a significant reduction of visual function that cannot be fully corrected by ordinary lenses, medical treatment, or surgery. Aging, injuries, and diseases can cause low vision. Its leading causes and those of blindness, include cataracts and age-related macular degeneration (AMD), both of which are more prevalent in the elderly population. Our overarching goal is to develop devices to aid people with low vision. We propose to use techniques of computer vision and computational neuroscience to build systems that enhance natural images. We plan test these systems on normal people and visually impaired older adults, with and without AMD. We have three milestones to reach: 1) Our first milestone will be to develop a system for low-noise image-contrast enhancement, which should help with AMD, because it causes lower contrast sensitivity. 2) Our second milestone will be a system that extracts the main contours of images in a cortical-like manner. Superimposing these contours on images should help with contrast-sensitivity problems at occlusion boundaries. Diffusing regions inside contours should help with crowding problems prominent in AMD. 3) Our third milestone is to probe whether these systems can help people with low vision people. For this purpose, we plan to use a battery of search and recognition psychophysical tests tailor-made for AMD. We put together an interdisciplinary team. The Principal Investigator is Dr. Norberto Grzywacz from Biomedical Engineering at USC. Drs. Gerard Medioni from Computer Science and Bartlett Mel from Biomedical Engineering at USC will lead the efforts in Specific Aims 1 and 2 respectively. Drs. Bosco Tjan from USC Psychology, Susana Chung from the University of Houston, and Eli Peli from Harvard Medical School will lead Specific Aim 3. Other scientists are from USC. They include Dr. Irving Biederman from Psychology, an object-recognition expert and Dr. Mark Humayun, from Ophthalmology, an AMD expert. They also include Dr. lone Fine, from Ophthalmology, an expert in people who recover vision after a prolonged period without it and Dr. Zhong-Lin Lu, an expert on motion perception and perceptual learning.