"New Approach to Detecting Dim Targets in Complex Environments Using Dynamic Computing"

The Rapid Advancement in Artificial Intelligence Systems and Sensing Technologies

The rapid advancement in artificial intelligence systems and sensing technologies in recent years has ushered in a new era of possibilities for detecting and tracking objects, animals, and people. Intelligent machine vision, which involves the accurate and automated detection of visual targets, has numerous applications ranging from security and surveillance to environmental monitoring and medical imaging analysis.

While machine vision has shown promising results, it often struggles in low light or limited visibility conditions. To effectively detect and track faint targets, it is crucial for these systems to be able to extract features such as edges and corners from images. However, traditional sensors based on complementary metal-oxide-semiconductor (CMOS) technology are often unable to capture these features.

In response to this challenge, researchers from Nanjing University and the Chinese Academy of Sciences have proposed a new approach to developing sensors that can better detect dim targets in complex environments. Their method, outlined in a recent paper in Nature Electronics, involves in-sensor dynamic computing, which combines sensing and processing capabilities in a single device.

Conventional Techniques for Detecting Targets in Images

Conventional techniques for detecting targets in images rely on CMOS-based sensors that operate independently. While some of these techniques have shown promise, they struggle to accurately distinguish between target signals and background signals. To address this issue, computer scientists have been exploring new principles for hardware development using low-dimensional materials that can be easily produced using established techniques and are also compatible with CMOS technology. The goal of this research is to achieve higher precision and robustness in low-contrast optical environments.

The Team Led by Feng Miao and Liang

The team led by Feng Miao and Liang introduced a new in-sensor dynamic computing approach that can detect and track dim targets in unfavorable lighting conditions. This method utilizes multi-terminal photoelectric devices based on graphene/germanium mixed-dimensional heterostructures, which are combined to create a single sensor.

What sets this approach apart is its use of dynamic feedback control between interconnected and neighboring optoelectronic devices based on multi-terminal mixed-dimensional heterostructures. Initial tests have shown that this method is highly effective in tracking dim targets in challenging lighting conditions.

Importantly, the devices used in this approach are made of graphene and germanium, materials that are compatible with CMOS technology and can be produced on a large scale. In the future, the team's approach could be tested in various real-world scenarios to further demonstrate its potential.

More Information

More information: Yuekun Yang et al. In-sensor dynamic computing for intelligent machine vision, Nature Electronics (2024). DOI: 10.1038/s41928-024-01124-0 Journal information: Nature Electronics © 2024 Science X Network

Ann Castro
Ann Castro Author
Ann Castro carries a total of 7 years experience in the healthcare domain. She owns a Master’s of Medicine Degree. She bagged numerous awards by contributing in the medical field with her ground-breaking notions. Ann has developed her own style of working and known for accuracy in her work. She loves trekking. She visits new places whenever she gets free time.