Neuromorphic vision sensor for real-time detection and tracking

用于实时检测和跟踪的神经形态视觉传感器

• 批准号: NI230100026
• 负责人: Prof Sumeet Walia
• 金额: $ 39.63万
• 依托单位: RMIT University
• 依托单位国家: 澳大利亚
• 项目类别: National Intelligence and Security Discovery Research Grants
• 财政年份: 2023
• 资助国家: 澳大利亚
• 起止时间: 2023-11-15 至 2026-11-14
• 项目状态: 未结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/NI230100026
• 关键词: Neuromorphic; vision; sensor; real; time

The project aims to develop a miniaturised sensory device for event perception inspired by human vision and brain. The project expects to utilise an interdisciplinary approach combining optically active materials, nanofabrication of a broadband light sensor with in-built memory, and spiking neural networks to create real-time, event-based detection and tracking capability while reducing redundant data and latency. The expected outcome is an autonomous vision device that highlights changes in the scene using visible and infrared wavelengths. This should provide significant benefits to the security, defence, intelligence and space sectors through integrated stealth detection and tracking of targets real-time even in poor lighting.

该项目旨在开发一种受人类视觉和大脑启发的事件感知传感器。该项目预计将利用一种跨学科的方法，将光学活性材料、具有内置存储器的宽带光传感器的纳米纤维和尖峰神经网络相结合，以创建基于事件的实时检测和跟踪能力，同时减少冗余数据和延迟。预期的结果是一个自主视觉设备，突出显示的变化，在现场使用可见光和红外波长。这将为安全、国防、情报和空间部门提供重大利益，因为即使在光线不好的情况下，也能对目标进行综合隐形探测和实时跟踪。