近日，谷歌在其官方博客上宣布，开源了名为Magika的先进技术，该技术能够实现毫秒级的文件格式和内容类型识别。这一突破性的进展，标志着人工智能在文件处理领域的应用又迈出了重要的一步。
Magika采用了一个定制的深度学习模型，该模型经过高度优化，即使在普通的CPU上运行，也能在几毫秒内准确识别出文件的类型。这一技术的应用，不仅能够极大提升文件处理的效率，还能在多种应用场景中发挥重要作用，如文件管理和安全检查等。
谷歌表示，Magika已经在百万级文件的测试中表现出色，识别准确率超过99%。这一开源项目的推出，无疑将促进全球开发者和研究人员之间的合作与交流，共同推动人工智能技术的发展。
此外，谷歌还表示，开源Magika的源代码已经托管在GitHub上，任何对文件识别技术感兴趣的开发者都可以访问和利用这些代码，进一步推动技术的创新和应用。
英文标题：Google Open-Sources Magika: Millisecond File Type Identification
英文关键词：Artificial Intelligence, File Recognition, Open Source
英文新闻内容：
Google has recently announced the open-source release of Magika, an advanced technology capable of identifying file formats and content types in milliseconds. This breakthrough marks a significant step forward in the application of artificial intelligence in file processing.
Magika utilizes a custom, highly optimized deep learning model that can accurately identify file types on a standard CPU within a few milliseconds. The application of this technology can greatly enhance the efficiency of file processing and play a crucial role in various application scenarios, such as file management and security checks.
Google has stated that Magika has demonstrated excellent performance in tests involving millions of files, with an accuracy rate exceeding 99%. The release of this open-source project will undoubtedly foster collaboration and communication among global developers and researchers, promoting the development of artificial intelligence technology.
Furthermore, Google has made the source code of Magika available on GitHub, allowing developers interested in file recognition technology to access and utilize these codes, further driving innovation and application in the field.
【来源】https://www.ithome.com/0/750/474.htm
Views: 1