Image/video restoration in ill lighting conditions

光照条件不佳时的图像/视频恢复

• 批准号: 517550-2017
• 负责人: Wu, Xiaolin
• 金额: $ 4.95万
• 依托单位: McMaster University
• 依托单位国家: 加拿大
• 项目类别: Collaborative Research and Development Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=719420
• 关键词: Image; video; restoration; ill; lighting

Matching the importance of vision to the human brain, digital images of all types and sizes are the dominant form of inputs and outputs in most intelligent devices and systems, such as smart phones, smartcars, smart cities, etc. The ubiquity of digital imageries is a landmark of modern technological societies and knowledge-based economies. In this project we will research on advanced algorithms to restore and enhance images/videos acquired in adverse conditions, such as low illumination, backlighting, camera motions, etc. The main objective is to make image quality exceed the quality (e.g., clarity, dynamic range, resolution) of raw semiconductor image sensors by sophisticated image processing and machine learning algorithms. Theresulting new techniques will be adopted by our industrial partner, Huawei Technologies Canada Co. , LTD., in their next generation products (e.g., smart phones) and cloud services. Through the McMaster-Huawei collaborations and the technology transfer, this NSERC CRD project will contribute to the the competitiveness of Canadian IT industry and the training of highly qualified personnel.

与人类大脑视觉的重要性相匹配，各种类型和大小的数字图像是大多数智能设备和系统（如智能手机，智能汽车，智能城市等）中输入和输出的主要形式。 在这个项目中，我们将研究先进的算法，以恢复和增强在不利条件下获取的图像/视频，如低照度，背光，相机运动等，主要目标是使图像质量超过质量（例如，通过复杂的图像处理和机器学习算法，我们可以对原始半导体图像传感器的清晰度、动态范围、分辨率等进行优化。 由此产生的新技术将被我们的工业合作伙伴华为技术加拿大有限公司采用，在他们的下一代产品中（例如，智能手机）和云服务。 通过McMaster-Huawei的合作和技术转让，该NSERC CRD项目将有助于提高加拿大IT行业的竞争力和培养高素质的人才。