麻豆淫院

Over the past decades, computer scientists have developed increasingly sophisticated sensors and machine learning algorithms that allow computer systems to process and interpret images and videos. This tech-powered capability, also referred to as machine vision, is proving to be highly advantageous for the manufacturing and production of food products, drinks, electronics, and various other goods.

Machine vision could enable the automation of various tedious steps in industry and manufacturing, such as the detection of defects, the inspection of electronics, automotive parts or other items, the verification of labels or expiration dates and the sorting of products into different categories.

While the sensors underpinning the functioning of many previously introduced machine vision systems are highly sophisticated, they typically do not process visual information with as much detail as the human retina (i.e., a light-sensitive tissue in the eye that processes visual signals).

Most sensors used so far are frame-based or event-based. Frame-based sensors are designed to capture images at specific intervals, while event-based sensors record changes in brightness at individual pixels. Notably, both these types of sensors are not yet as fast and adaptable as the human retina.

Researchers at the Chinese Academy of Sciences, the Sino-Danish Center for Education and Research and other institutes recently developed a sensing device that mirrors the retina's layered structure. Their proposed device, introduced in a paper in Nature Nanotechnology, is a photodiode (PD), a semiconductor device that can convert light into an electrical current.

"Current machine vision sensors, including frame-based and event-based types, often fall short due to their limited temporal dynamics compared with the human retina, hindering their overall performance and adaptability," wrote Qijie Lin, Congqi Li and their colleagues in their paper.

"We present an event-driven retinomorphic photodiode (RPD) that mimics the retina's layered structure and signal pathway. The RPD achieves this by vertically integrating an organic donor鈥揳cceptor heterojunction, an ion reservoir with a porous web-like morphology, and a Schottky junction into a single diode through controlled layer-by-layer fabrication and precise nanostructure modulation."

The retina-inspired device developed by the researchers has three key components. These are an organic donor-acceptor heterojunction, an ion reservoir made of porous nanostructures and a Schottky junction.

The first is essentially a junction of two organic semiconductors that facilitates the transfer of electrical charge. The ion reservoir is a sponge-like structure that can store and release ions, ultimately mimicking how biological tissues use ions to carry signals. Finally, the Schottky junction is an interface formed between a semiconductor and metal that allows electric current to flow smoothly in one direction but not in the other.

"Each component replicates a key retinal process, and their spontaneous interaction results in environment-adaptive dynamics," wrote Lin, Li and their colleagues. "This design yields a dynamic range exceeding 200鈥塪B, substantially reduces noise and data redundancy, and allows for high-density integration. We demonstrate that these improvements enable high-quality machine vision, even under extreme lighting conditions."

In initial tests, the new RPD developed by Lin, Li and their colleagues achieved highly promising results, outperforming other PD previously tested in machine vision tasks. In the future, the new device could be improved further and applied to a broad range of real-world machine vision tasks.

"Our work demonstrates a bottom-up approach to retinomorphic sensors, propelling the development of robust and responsive machine vision systems adaptable to complex and dynamic lighting environments," wrote the authors.

漏 2025 Science X Network