Researchers at UCLA have developed a lens-free, wide-field super-resolution imaging platform that uses a scanned illumination aperture and computational reconstruction to overcome the pixel size limit and achieve high-resolution imaging across large fields of view. The system captures multiple low-resolution holograms under shifted illumination and uses sub-pixel registration and algorithmic reconstruction to produce a high-resolution final image.
Traditional optical microscopy systems often face a trade-off between field of view (FOV) and spatial resolution. When placing the sample very close to the sensor (on-chip holography), pixel size of the image sensor becomes a limiting factor, preventing resolution enhancement. At the same time, using magnification optics reduces the FOV and introduces complexity. To bring super-resolution imaging to compact, wide-area, lens-free systems, novel approaches are needed to circumvent the pixel-size constraint computationally, while maintaining large fields of view.
The system uses a scannable illumination aperture (e.g. a LED or laser source through a small aperture) that is raster scanned over multiple positions relative to the sample.
Each scan position yields a lower-resolution hologram captured directly on the sensor, without lenses.
Because the illumination shift corresponds to sub-pixel shifts at the sensor plane, the method recovers high spatial frequency information by combining these multiple holograms (i.e. a super-resolution reconstruction algorithm) to surpass the native sensor pixel limit.
The algorithm includes self-calibration: the system does not require external knowledge of the scanning step. The shifts are inferred during reconstruction from raw holograms, simplifying hardware alignment and making the system robust.
Because the illumination is partially coherent and from a relatively large aperture, the system suppresses speckle noise and interference artifacts compared to coherent-holography approaches.
The reconstruction yields a numerical aperture ~0.5, enabling ~0.6 μm spatial resolution at visible wavelengths, while retaining a large imaging FOV (e.g. ~24 mm²) corresponding to the full detector area.
Achieves high spatial resolution (sub-micron) even when limited by sensor pixel size.
Maintains large field of view: FOV is equal to the active area of the detector, rather than reduced by magnification optics.
Hardware simplicity: no refractive optics, lenses, or complex alignments required.
Self-calibrating shift estimation reduces mechanical complexity and calibration burden.
Reduced noise artifacts due to partially coherent illumination and large aperture usage.
Scalable: larger sensor chips yield proportionally larger FOV at high resolution.
On-chip microscopy for cell biology: imaging live cells, pathogens, or microstructures over large fields.
Portable diagnostic imaging systems, e.g. point-of-care microscopy.
High-throughput screening platforms requiring wide-area imaging with sub-micron details.
Environmental imaging (microplastics, particulates, organisms) in situ.
Lab-on-chip devices combining imaging and microfluidics.
Low-cost imaging devices for resource-limited settings (e.g. in field diagnostics).
US 8,866,063 B2 — Lens-Free Wide-Field Super-Resolution Imaging Device Lens-Free Wide-Field Super-Resolution Imaging Device (US8866063B2) Google Patents
Bishara, W.; Rotella, C.; Ozcan, A. “Lens-free, pixel super-resolution, wide-field on-chip microscopy via self-assembled masks.” Optics Express 18(11): 11181-11191 (2010). DOI: 10.1364/OE.18.011181 — describes foundational techniques in mask scanning and computational super-resolution for lens-free imaging.