Far-Field Super-Resolution Imaging Using Limited-Size Object Microscopy

Dr Chang Taeyong (left) with the super-resolution microscope developed for this research work, Dr Giorgio Adamo (top right) and Professor Nikolay I. Zheludev (bottom right). 

For more than a century, scientists believed there was a limit to how well small objects could be observed. In the late 19^th century, scientists such as Ernst Abbe, Hermann von Helmholtz and Lord Rayleigh demonstrated that light itself imposes a limit on how clearly small objects can be imaged. This became known as the diffraction limit.

Diffraction occurs because light behaves as a wave and spreads out after interacting with an object. Because of this spreading, even an ideal optical system cannot distinguish details separated by less than a certain distance, typically about half the wavelength of light (λ/2). This principle has long served as a benchmark for conventional microscopy and optical systems design. However, achieving label-free, far-field super-resolution imaging (resolving features smaller than the diffraction limit without placing probes near the sample or using fluorescent labels) has remained a major challenge in modern optics.

Researchers at Nanyang Technological University (NTU), Singapore, and the University of Southampton in the United Kingdom, have now developed a method called limited-size object microscopy (LSOM). This technique takes advantage of prior knowledge that the object being imaged is confined within a limited space. Using an advanced tool known as Slepian-Pollak functions, or prolate spheroidal wavefunctions, the method reconstructs fine structural details from diffraction-limited measurements. Without relying on fluorescent labels or near-field probes, the researchers achieved resolutions significantly beyond the traditional limit, demonstrating λ/7 resolution for two-dimensional objects and λ/8 resolution for one-dimensional objects.

This study, titled “Super-resolution imaging of limited-size objects,” was published in the February 2026 issue of Nature Photonics. The lead author on the paper, Dr Taeyong Chang, is a postdoctoral researcher at the School of Physical and Mathematical Sciences (SPMS) at NTU. The project was supervised by Dr Giorgio Adamo, a senior principal research fellow at SPMS, and Prof. Nikolay Zheludev.

Nanoscale Imaging without Fluorescent Labels or Near-field Probes

Early studies in optical theory suggested that reconstructing details beyond the diffraction limit from conventional optical measurements would be practically impossible due to the extremely high precision required to overcome noise. The new LSOM approach overcomes this limitation by enabling accurate imaging of nanoscale objects without needing fluorescent labels, near-field probes, or prior assumptions about the object’s shape. “We were motivated by the limitations of existing super‑resolution techniques, which often rely on invasive probes, staining agents, or highly specialized conditions,” says Dr Giorgio Adamo. “Our goal was to develop a universally applicable, non‑invasive method capable of achieving deep subwavelength resolution using only far‑field scattered light.

Figure 1. (a) In superoscillatory hotspot generation, a carefully designed pattern in the Fourier plane produces a wave field that creates subwavelength features within a limited region on the object plane. (b) In limited-size object microscopy, this process is reversed: light scattered from a spatially confined object produces a band-limited field in the Fourier plane, which is analyzed using Slepian-Pollak coefficients to reconstruct the object’s fine details.

The concept behind LSOM can be understood as the reverse process of superoscillatory hotspots generation, where interfering waves produce features smaller than the conventional diffraction limit. In the LSOM apparatus, a digital micromirror device selectively isolates specific optical modes, while a shot-noise-limited interferometric detection scheme enables highly precise measurements. The team also found that careful system calibration of the imaging system is essential to reconstruct the object accurately and compensate for optical distortions. Using this approach, the researchers successfully imaged platinum and gold nanoparticles with high resolution and strong robustness against noise, whereas conventional optical imaging methods could not resolve these nanoscale features.

Figure 2. Experimental demonstration of LSOM based on scanning electron microscopy (SEM) images of nanoscale objects with a range of shapes, sizes and symmetries. The LSOM images closely match both the actual electron microscope images and computer simulations, showing much shaper detail. Overall, the results demonstrate a major improvement in resolution, equivalent to a much higher effective microscope strength. 

Expanding Applications of LSOM: From Nanoscale Particle to Biological Imaging

The team analysed LSOM using information theory, treating the microscope as a communication channel that transferring data from the object to the detector. Their framework reveals that performance depends on a trade-off between three factors: target resolution, object size and photon budget (the number of photons detected). As a result, smaller objects can be reconstructed at a higher resolution for a given number of photons. This analysis established the measurement precision and experimental conditions required to achieve super-resolution beyond the conventional diffraction limit.  

Although the technique is currently limited to small imaging areas, it is applicable to systems where objects are isolated or clustered within a confined region. This includes identifying nanoparticles and nanowires in nanotechnology, or microscopic pollutant particles in environmental monitoring. Beyond microscopy, LSOM may also benefit fields like precision metrology, optical spectroscopy and light detection and ranging (LiDAR), where accurate reconstruction of optical fields is essential.

“Label-free far-field super-resolution imaging has long been a major challenge in optical physics. Achieving it experimentally, without prior knowledge of an object’s structure beyond its finite extent, requires extremely precise measurements and an optical apparatus designed to reduce noise to an absolute minimum,” says Dr Taeyong Chang. Dr Giorgio Adamo adds that “LSOM demonstrates that far-field, label-free super-resolution imaging is possible when an object is confined to a limited spatial region. This technique effectively complements existing imaging methods and opens the door to high-resolution imaging of viruses, environmental nanoparticles, and technologically relevant nanostructures like quantum dots.”

The researchers are currently collaborating with biologists at Nanyang Technological University to apply LSOM to image bacteriophages, marking an exciting leap toward biological and biomedical applications. Their long-term vision is for this microscope to become a standardized tool in research laboratories, clinical diagnostics and industrial nanotechnology characterization settings.