Integrated Photonic Processor Implementing Digital Image Convolution

• DOI: 10.3390/electronics14040709
• Publication Date: 2025-02-12T15:25:57Z
• Authors: Chensheng Wang; Wenhao Wu; Zhenhua Wang; Zhijie Zhang; Wei Xiong; Leimin Deng
• Keywords: Convolution (computer science); Image processor

Upon the advent of big data era, information processing hardware platforms have undergone explosive development, facilitating unprecedented computational capabilities while significantly reducing energy consumption. However, conventional electronic computing hardware, despite significant upgrades in architecture optimization and chip scaling, still faces fundamental limitations speed efficiency due to Joule heating, electromagnetic crosstalk, capacitance. A new type is urgently needed for emerging data-intensive applications such as face identification, target tracking, autonomous driving. Recently, integrated photonics architecture, which possesses remarkable compactness, wide bandwidth, low latency, inherent parallelism, has harvested great attention its enormous potential accelerate parallel processing, digital image convolution. In this study, an photonic processor based on a Mach-Zehnder interferometer (MZI) network proposed demonstrated. The processor, being scalable compatible with complementary metal oxide semiconductors, facilitates mass production seamless integration other silicon-based optoelectronic devices. An experimental verification convolution also performed, result deviations between our commercial 64-bit computer are less than 2.3%.