With the advantages of parallel processing, high-speed operation, and ultra-low power consumption, optical computing presents immense potential in this age of information explosion. As a fundamental optical computing component, optical differentiator has extensive applications, particularly excelling in the real-time processing and recognition of high-frame-rate image information. An ideal optical differentiator needs to adapt to different application scenarios, such as arbitrary operating wavelengths, complex or even transparent targets. Furthermore, performing differential operations of varying orders on target optical field can extract multi-level information. Unfortunately, most current differential computing devices can only meet part of the requirements, which limits the attained information dimensions and hinders the development of their practical application. Herein, a universal platform based on liquid crystals (LCs) is proposed to achieve 0th-, 1st-, and 2nd-order differential operations of amplitude-typed or even phase-typed objects within a broadband range. The switching of different differentiation orders can be accomplished by straightforward adjusting the applied voltages of the differentiator. This approach is capable of extracting more feature information including bright-field, outline, and edges images of versatile targets. With advanced characteristics of multi-dimension, broadband and multifunction, the work can flourish applications in fields of image processing, biomedical imaging and optical analog computation, etc.

中文翻译：

---

基于动态液晶的全光多阶复用微分

光计算凭借并行处理、高速运行、超低功耗等优势，在信息爆炸的时代展现出巨大的潜力。光学微分器作为一种基础的光学计算器件，有着广泛的应用，特别是在高帧率图像信息的实时处理和识别方面表现出色。理想的光学微分器需要适应不同的应用场景，例如任意工作波长、复杂甚至透明的目标。此外，对目标光场进行不同阶次的微分运算可以提取多层次的信息。遗憾的是，目前大多数差分计算装置只能满足部分要求，限制了所获得的信息维度，阻碍了其实际应用的发展。在此，提出了一种基于液晶（LC）的通用平台，以在宽带范围内实现幅度型甚至相位型物体的0阶、1阶和2阶微分运算。不同微分阶数的切换可以通过直接调整微分器的施加电压来完成。这种方法能够提取更多的特征信息，包括多功能目标的明场、轮廓和边缘图像。该工作具有多维、宽带、多功能的先进特点，可在图像处理、生物医学成像、光学模拟计算等领域得到广泛应用。