Observing and recording transient events is crucial for understanding their fundamental physical principles and for enabling control over related processes, such as inertial confinement fusion, laser-material interactions, plasma physics, and laser surgery. To investigate the intrinsic mechanisms of these transient phenomena, an ultrafast imaging technique with high spatiotemporal resolution is necessary. Although high-speed cameras can capture dynamic processes at millions of frames per second, which meets the observational needs at the microsecond scale, their ability to record shorter timescale dynamics is constrained by the sensor's readout speed. In contrast, framing imaging techniques employ spatiotemporal segmentation to record transient events using multiple physical or virtual cameras, thereby overcoming the limitations imposed by sensor readout speed and achieving higher temporal resolution. Consequently, these techniques have become primary tools for studying ultrafast phenomena. However, existing optical framing imaging methods often face challenges such as limited sequence depth, low image quality, inadequate temporal resolution, or fixed inter-frame intervals. These limitations impede the accurate detection of ultrafast dynamic processes.

To address these challenges, Professor Shian Zhang's team at East China Normal University has developed a spatiotemporal shearing-based ultrafast framing imaging technique, providing a high-performance solution for transient imaging of ultrafast scenes occurring on timescales ranging from sub-nanoseconds to nanoseconds. Relevant research results were recently published in Photonics Research, Volume 13, Issue 3, 2025. [Yu He, Yunhua Yao, Jiali Yao, Zhengqi Huang, Mengdi Guo, Bozhang Cheng, Hongmei Ma, Dalong Qi, Yuecheng Shen, Lianzhong Deng, Zhiyong Wang, Jian Wu, Zhenrong Sun, Shian Zhang, "Spatiotemporal shearing-based ultrafast framing photography for high performance transient imaging," Photonics Res. 13, 642 (2025)]

This study proposes a novel approach that integrates discrete pulse train illumination with spatiotemporal shearing imaging. The experimental setup is illustrated in Fig. 1. The STS-UFP system utilizes a pulse train generation device based on spectrum shuttle to produce illumination pulses. This device allows for flexible control over the number and temporal spacing of sub-pulses, enabling adjustments to the imaging sequence depth and time window for discrete sampling of ultrafast dynamic scenes. Furthermore, the sub-pulses within the pulse train have ultrashort durations, facilitating precise temporal slicing of dynamic scenes and mitigating motion blur caused by spatiotemporal aliasing. A streak camera captures the dynamic scene through spatiotemporal shearing, redistributing images from different temporal moments to distinct spatial positions. To balance the trade-off between field of view (FoV) and imaging sequence depth in the STS-UFP system, a spatial slicing module is introduced. This module duplicates the dynamic scene and redistributes it horizontally while introducing a vertical offset. After passing through the entrance slit of the streak camera, the recorded region is transformed into three horizontally arranged slices. This approach effectively optimizes the balance between imaging sequence depth and FoV height, enhancing the overall imaging performance of the system. The time window of the STS-UFP system can be flexibly adjusted by modifying the pulse train interval and the slope of the streak camera's scanning voltage, accommodating different temporal detection requirements. Finally, image processing algorithms are employed to reconstruct a discrete sequence of dynamic scene images. To improve image quality, Laplacian blending is applied to minimize intensity mismatches at stitching locations, while a DRU-Net-based denoising algorithm is incorporated to effectively suppress noise, thereby optimizing imaging performance. The data processing workflow is depicted in Fig. 2.

Figure 1. Schematic of the STS-UFP experimental system.

Figure 2. Schematic diagram of the data processing workflow in the STS-UFP system.

STS-UFP enables high-fidelity framing imaging with a sequence depth of up to 16 frames. The frame interval can be adjusted within a range of hundreds of picoseconds to nanoseconds, while maintaining exposure times at the picosecond level. The STS-UFP system has been utilized to observe plasma and shock waves induced by femtosecond lasers in water, the ablation process of biological tissue by femtosecond lasers, and the shock waves generated on silicon surfaces by femtosecond lasers.

As shown in Figure 3, the femtosecond laser, tightly focused by the objective lens, induces self-focusing near the focal point, rapidly ionizing molecules and forming a plasma channel. In the image taken at the 4 ns time mark, a shock wave layer is observed outside the plasma channel, with two red triangles indicating the boundary between the shock wave and the plasma channel. After 4 ns, the outlines of the shock wave and plasma become increasingly distinct and expand outward from the plasma channel.

Figure 3. (a) Experimental setup for observing femtosecond laser-induced plasma and shock wave expansion in water using STS-UFP. (b) Processed image of plasma and shock wave expansion induced by a 400 nm femtosecond laser pulse in water. (c) Expansion curve of plasma and shock wave.

As shown in Figure 4, the image at 0 ns reveals the formation of an optical filament after focusing. When the femtosecond laser irradiates the silicon surface, images between 0.5 to 1.5 ns show the generation of a large amount of plasma in the ablation region, which appears black in the image. This is because the excited electrons in the plasma absorb the probing laser. Subsequently, the plasma becomes gradually transparent and pushes the surrounding air outward, ultimately forming a shock wave. Between 2 to 7 ns, the shock wave front rapidly expands and shows a tendency to further expand.

Figure 4. (a) Schematic diagram of the experimental setup for observing femtosecond laser-induced shock waves on a silicon surface using STS-UFP. (b) Processed image of the femtosecond laser-induced shock wave. (c) The propagation distances of the shock wave in the horizontal and diagonal 45° directions were extracted and analyzed. Additionally, the propagation distance in the diagonal 45° direction was fitted using the Sedov-Taylor theory.

Professor Xianfeng Chen commented that, this technique combines time-domain discrete illumination with spatio-temporal shearing imaging, achieving large sequence depth, high image quality, ultra-short exposure time and adjustable frame intervals. It provides a solution for high-fidelity long-sequence in-situ ultrafast observation, offering a powerful tool for further research on the fine processes of ultrafast phenomena, such as exploring femtosecond laser ablation mechanisms, optimizing laser processing parameters and guiding laser surgery.