OBJECT RECOGNITION & NAVIGATION--NORMAL & LOW VISION

物体识别

• 批准号: 6384301
• 负责人: GORDON E LEGGE
• 金额: $ 22.07万
• 依托单位: UNIVERSITY OF MINNESOTA
• 依托单位国家: 美国
• 财政年份: 1978
• 资助国家: 美国
• 起止时间: 1978-12-01 至 2004-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6384301
• 关键词: behavioral /social science research tag; chordate locomotion; computer simulation; eye movements; form /pattern perception; human subject; memory; neural information processing; psychophysics; saccades; vision aid; vision disorders; visual feedback; visual fields; visual perception; visual threshold

DESCRIPTION (Taken from the applicant's Abstract):There are 3.5 million Americans with low vision. Many have daily problems navigating and recognizing objects around them. These problems can threaten personal independence and quality of life. There has been almost no research on low-vision object recognition or navigation. Theoretical and experimental studies in this proposal have the broad goal of understanding and ameliorating these problems. Most people with low vision contend with either visual-field loss or field restriction form magnifiers. As a result, viewing of objects or environments often requires a piecemeal search of the scene for informative details. People with normal vision use eye movements to explore scenes, especially when visibility is poor. The major theoretical goal is to develop a conceptual framework for describing visual exploration of objects and environments. We will develop Active ideal-observer models of object recognition and navigation. These models will perform optimally in a given visual task, and will serve as benchmarks for interpreting the performance of people with normal and low vision. One of the experimental goals is to study eye-movement exploration and magnifier exploration to recognize objects. What visual features and cognitive strategies guide this exploration? A second experimental goal is to determine whether training on cognitive maps can be a useful navigation aid for people with low vision. (Cognitive maps are memory representations of visual and spatial information about environments.) The team will perform psychophysical experiments in virtual and real buildings. The long-term goal is to develop a computer-based method for imparting information about building layouts to visually disabled people. In this grant, the team will study the impact of impaired vision, esp. field loss, on the formation and use of cognitive maps. Their hypothesis is that cognitive maps, although difficult to learn with visual impairment, will be esp. beneficial to people with low vision.

描述(摘自申请者摘要)：有350万 视力低下的美国人。许多人每天都有导航和 辨认周围的物体。这些问题可能会威胁到个人 独立性和生活质量。几乎没有关于这方面的研究 低视力物体识别或导航。理论上和 这项建议中的实验研究的广泛目标是 认识和改进这些问题。大多数人患有低血压病 视力与视野损失或视野限制形式相抗衡 放大镜。因此，查看对象或环境通常 需要对现场进行零星搜索，以获取信息细节。人民 视力正常的人使用眼球运动来探索场景，特别是当 能见度很低。主要的理论目标是开发一种 用于描述对象的可视探索的概念框架和 环境。我们将开发对象的主动理想观察者模型 识别和导航。这些型号将在 给出视觉任务，并将作为解释 视力正常和低视力者的表现。其中一个 实验目标是研究眼动探索和放大镜 探索以识别物体。什么视觉特征和认知 战略指导这一探索？第二个实验目标是 确定认知地图上的训练是否可以成为有用的导航 为视力低下的人提供援助。(认知图是记忆 关于环境的视觉和空间信息的表示。) 该团队将在虚拟和真实中进行心理物理实验 建筑物。长期目标是开发一种基于计算机的方法来 将有关建筑物布局的信息传递给视觉上禁用的人 人民。在这笔赠款中，该团队将研究受损害的 远见(尤指)场损失，对认知图谱的形成和使用。 他们的假设是，认知地图，尽管很难学习 有视力障碍的，将特别是。有益于低收入人群 幻象。