Revolutionizing imaging: programmable lensless holographic camera

Traditional cameras rely on lenses to focus light and image sensors to record it. While advances in consumer electronics have enabled the miniaturization and cost reduction of image sensors, the size, weight, and manufacturing complexity of high-performance lenses remain major bottlenecks—particularly in applications that demand extreme portability, such as augmented reality (AR), virtual reality (VR), and wearable devices. Lensless imaging offers a promising alternative by eliminating the bulky lens. Instead, it uses a lightweight patterned mask to modulate incoming light, followed by computational reconstruction of the image. However, most existing lensless systems employ fixed masks, which limits their adaptability to diverse scenes. This often results in aliasing artifacts and unstable reconstructions under non-ideal conditions, constraining image quality and system versatility.

To overcome these challenges, researchers from the Smart Computational Imaging Laboratory (SCILab) at Nanjing University of Science and Technology, led by Professor Chao Zuo, have developed a programmable lensless imaging system based on a variable Fresnel zone aperture (FZA) displayed on an LCD screen. Their recent work was published in Volume 11, Issue 12 of Science Advances on 21 March 2025. In this study, they introduce a new imaging framework called LIP (Lensless Imaging with a Programmable FZA). Rather than relying on a static pattern, the LIP system dynamically adjusts the mask in real time to capture richer information across multiple dimensions. These measurements are then synthesized using computational algorithms to reconstruct high-resolution images with an enhanced signal-to-noise ratio (SNR).

Elaborating on the findings, Prof. Zuo says, “This space-frequency joint framework supports adaptive switching of encoding strategies for static or dynamic scenes, enabling real-time imaging with both high resolution and frame rate.”

Using a self-developed, palm-sized, miniaturized prototype, the research team conducted a series of experiments to evaluate the system’s performance. Compared to conventional static-modulation lensless imaging methods, LIP demonstrated a 2.5× improvement in spatial resolution and a 3 dB enhancement in SNR. In dynamic gesture recognition experiments—including actions such as tapping, zooming, and rotating—the LIP system maintained real-time reconstruction at 15 frames per second (fps), all while being nearly 90% smaller than traditional gesture-sensing devices.

Prof. Zuo briefly explains the technological impact of this new imaging system. “LIP is poised for broader impact through integration with emerging technologies such as intelligent algorithms, novel spatial light modulators, and advanced microfabrication.”

They also outlined several future directions for LIP. One promising avenue is multi-modal sensing, where LIP is integrated with polarization, depth, or spectral modalities to provide richer optical information for biomedical imaging and material analysis. Another key focus is portable applications, aiming to develop compact, lightweight, and high-performance imaging systems suitable for wearables, augmented reality (AR), virtual reality (VR), and mobile platforms.

Additionally, focal plane encoding represents an exciting direction, extending the programmable architecture directly to the image sensor level. This approach would enable each pixel to capture not only intensity but also directional or wavefront information, significantly enhancing imaging capabilities.

“These developments may usher in the next generation of compact, high-performance computational imaging systems,” the research team noted, “opening new possibilities in science, industry, and daily life.”

***

Reference

Authors: Xu Zhang^1,2,3, Bowen Wang^1,2,3, Sheng Li^1,2,3, Kunyao Liang^1,2,3, Haitao Guan^1,2,3, Qian Chen^1,2,3, Chao Zuo^1,2,3

Title of original paper: Lensless imaging with a programmable Fresnel zone aperture

Journal: Science Advances

DOI: 10.1126/sciadv.adt3909

Affiliations         

¹School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing, China.

²Jiangsu Key Laboratory of Spectral Imaging & Intelligent Sense, Nanjing, China.

³Smart Computational Imaging Laboratory (SCILab), Nanjing University of Science and Technology, Nanjing, China.

About The Smart Computational Imaging Laboratory in Nanjing University of Science and Technology, China

The Smart Computational Imaging Laboratory (SCILab: www.scilaboratory.com), led by Professor Chao Zuo, is affiliated to the “Spectral Imaging and Information Processing” Innovation Team of the Ministry of Education's Changjiang Scholar and the first batch of “National Huang Danian-style Teacher Team”, led by Professor Qian Chen, the leader of the national first-level key discipline of optical engineering at Nanjing University of Science and Technology. The laboratory is committed to developing a new generation of computational imaging and sensing technologies. The research results have been published in more than 270 SCI journals, of which 46 papers were selected as cover papers of journals such as Light, Optica, AP, PhotoniX, etc., 25 papers were selected as ESI highly cited/hot papers, and the papers have been cited nearly 20,000 times.

About Professor Chao Zuo from Nanjing University of Science and Technology, China

Professor Chao Zuo, the academic leader of the Smart Computational Imaging Laboratory, is a distinguished professor of the Ministry of Education's Changjiang Scholars Program, a Fellow of SPIE | Optica | IOP, and is selected as a Clarivate Analytics Highly Cited Scientist in the World. He has won the Fresnel Prize of the European Physical Society, the first prize of the Technological Invention Award of the Chinese Society of Optical Engineering, the first prize of the Basic Category of the Jiangsu Science and Technology Award, and the “Special Commendation Gold Award” of the Geneva International Invention Exhibition.