**上海理工大学创新光学技术，卷积全光神经网络实现革命性成像突破**
近日，上海理工大学研究团队在光学与人工智能交叉领域取得了重大突破，成功开发出超快卷积光学神经网络（ONN），将卷积神经网络（CNN）的概念创新性引入光学领域。该技术的核心成果已发表于国际顶尖学术期刊《Science Advances》。
长久以来，CNN凭借其出色的图像识别能力在人工智能领域大放异彩。上海理工大学的研究团队通过独特的创新，将这一概念带入全新的光学领域，彻底改变了人工智能成像技术的格局。此次开发的超快卷积光学神经网络无需依赖光学记忆效应，便能实现对散射介质后方物体的高效清晰成像。其关键在于利用强散射过程直接提取特征，实现了无记忆图像重建的奇迹，将视野扩大了271倍，计算速度更是达到了惊人的每秒1.57千万亿次运算。
这一技术突破不仅在学术界引起了巨大关注，也在工业界和科研领域激发了热烈讨论。专家表示，该技术的实用化将为各种应用带来前所未有的高效成像能力，包括但不限于医疗诊断、安全监控、自动驾驶等领域。未来，上海理工大学的这一创新技术有望进一步推动全球光学和人工智能的融合与发展。
该研究标志着光学神经网络技术在图像处理和模式识别领域的重大进步，也预示着未来更多创新技术在智能成像领域的广泛应用。上海理工大学的研究团队正引领着一场光学革命，让世界见证了中国科研实力的崛起。
英语如下：
News Title: “Shanghai Polytechnic University Innovates in Optics: Convolutional Neural Network Accelerates Imaging Technology, Expanding Vision by 271 Times”
News Content: **Shanghai Polytechnic University Makes Breakthrough in Optical Technology, Revolutionizing Imaging with Convolutional All-Optical Neural Network**
Recently, a research team from Shanghai Polytechnic University has made significant progress in the intersection of optics and artificial intelligence, successfully developing an ultra-fast convolutional optical neural network (ONN) by innovatively introducing the concept of convolutional neural networks (CNN) into the field of optics. The core achievements of this technology have been published in the international top academic journal “Science Advances”.
For a long time, CNN has excelled in the field of artificial intelligence with its outstanding image recognition capabilities. Through unique innovations, the research team at Shanghai Polytechnic University has brought this concept into a new realm of optics, completely changing the landscape of artificial intelligence imaging technology. The ultra-fast convolutional optical neural network developed this time can achieve efficient and clear imaging of objects behind scattering media without relying on optical memory effects. The key is to extract features directly using a strong scattering process, achieving the miracle of memory-free image reconstruction, expanding the field of view by 271 times and achieving a stunning computational speed of 157 million trillion operations per second.
This technological breakthrough has not only attracted enormous attention in the academic community but also sparked lively discussions in the industrial sector and scientific research. Experts indicate that the practical application of this technology will bring unprecedented high-efficiency imaging capabilities to various applications, including but not limited to medical diagnosis, security monitoring, autonomous driving, and other fields. In the future, Shanghai Polytechnic University’s innovative technology is expected to further promote the integration and development of global optics and artificial intelligence.
This research marks a significant advancement in the field of image processing and pattern recognition with optical neural network technology, and it also预示未来更多创新技术在智能成像领域的广泛应用。上海理工大学的研究团队正引领着一场光学革命，让世界见证了中国科研实力的崛起。预示着未来更多创新技术在智能成像领域的广泛应用。The research team from Shanghai Polytechnic University is leading an optical revolution, demonstrating the rise of China’s scientific research capabilities to the world.
【来源】https://www.jiqizhixin.com/articles/2024-07-03-2
Views: 6