用二维材料光电二极管实现高分辨率单像素成像,突破此前64×64像素限制。
High Pixel Resolution Visible to Extended Shortwave Infrared Single Pixel Imaging with a black Phosphorus-Molybdenum disulfide (bP-MoS2) photodiode
- 采用bP-MoS2光电二极管与压缩采样算法,实现1023×768可见光与512×512红外成像。
- 通过循环S矩阵设计,将成像时间减少四分之一,同时保留关键图像特征。
- 方法简单易行,适合低成本红外成像系统,可应用于生物医疗与自动驾驶领域。
当前高分辨率红外成像仪因昂贵的传感器阵列而成本高昂。范德华(vdWs)材料为低成本、室温红外探测器提供了可能。尽管基于vdWs材料的探测器性能优异,但尚未实现兆像素阵列。单个vdWs探测器的成像通常依赖耗时的机械扫描,且分辨率较低。单像素成像(SPI)利用单一探测器与空间光调制器,可实现高分辨率成像。此前基于vdWs材料的SPI仅在近红外区实现过64×64像素的演示。本文首次展示基于黑磷-二硫化钼(bP-MoS₂)光电二极管的高分辨率SPI系统,实现可见光1023×768和扩展短波红外512×512像素的成像,像素数相较以往提升64倍。我们提出一种快速边缘检测方法用于特征提取,并采用基于哈达玛序列的循环S矩阵进行压缩采样,通过循环卷积与傅里叶变换实现高效重建,使测量次数减少,成像速度提升四倍。该方法为低成本短波与中波红外相机提供新路径,有望推动气体检测、生物医学成像、自动驾驶、安防监控等领域发展。
原文摘要 · Abstract (English)
High-resolution infrared imagers are currently more expensive than CMOS and CCD cameras, due to costly sensor arrays. Van der Waals (vdWs) materials present an opportunity for low-cost, room temperature infrared photodetectors. Although photodetectors based on vdWs materials show promising performance, creating a megapixel array is yet to be achieved. Imaging with a single vdWs photodetector typically relies on time-consuming mechanical scanning and suffers from low resolution. Single pixel imaging (SPI) offers an affordable alternative to achieve high-resolution imaging, utilizing only one photodetector and a spatial light modulator. Progress in SPI using vdWs material photodetectors has been limited, with only one prior demonstration in the near infrared range (64$\times$64 pixels). In this work, we demonstrate a high-resolution SPI system (1023$\times$768 for visible light and 512$\times$512 for extended shortwave infrared) using a black phosphorus-molybdenum disulfide (bP-MoS$_2$) photodiode, surpassing earlier vdWs material SPI implementations by a factor of 64 in pixel count. We introduce an easy-to-implement edge detection method for rapid feature extraction. We employ compressed sampling and reduce imaging time by a factor of four. Our compressed sampling approach is based on a cyclic S-matrix, which is derived from a Hadamard-based sequence, where each row is a circular shift of the first row. This enables efficient imaging reconstruction via circular convolution and Fourier transforms, allowing fewer measurements while preserving the key image features. Our method for SPI using a vdWs material photodetector presents the opportunity for inexpensive shortwave infrared and midwave infrared cameras, and thus may enable advances in gas detection, biomedical imaging, autonomous driving, security, and surveillance.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。