Breakthrough in Nanotechnology: Researchers Develop Reconfigurable Bioinspired Sensor for High-Accuracy Human Action Recognition

Researchers at the Institute of Materials Research and Engineering in Singapore have made a significant breakthrough in nanotechnology by developing a reconfigurable bioinspired sensor using ultrathin GaN/AlN-based quantum-disks-in-nanowires (QD-NWs) array. This innovative device has the potential to revolutionize computer vision and artificial intelligence by mimicking the human retina's Parvo and Magno cells. According to the research, the sensor's ability to recognize human actions has been significantly enhanced, with a recognition accuracy increasing from 51.4% to 81.4%.

Key Takeaways:

  • The researchers propose a bioinspired vision sensor composed of a GaN/AlN-based ultrathin QD-NWs array to mimic the human retina's Parvo and Magno cells.
  • The sensor's reconfigurability allows it to achieve two biosimilar photoresponse characteristics with slow and fast reactions to light stimuli, enhancing in-sensor image quality and HAR efficiency.
  • The interplay and synergistic interaction of the two photoresponse modes within a single device increase the HAR recognition accuracy from 51.4% to 81.4%.
  • The research demonstrates the potential of the intelligent vision sensor for developing highly efficient HAR systems and future smart optoelectronics.
  • Xin Ju, the lead researcher, notes that the development of such devices will enable more efficient and practical in-sensor vision computing.
  • The research team includes Zhixiang Gao, Huabin Yu, Wei Chen, Xin Liu, Yuanmin Luo, Yang Kang, Dongyang Luo, JiKai Yao, Wengang Gu, Muhammad Hunain Memon, Yong Yan, and Haiding Sun.

Statistics:

  • The recognition accuracy of the bioinspired sensor increases from 51.4% to 81.4% due to the interplay of the two photoresponse modes.
  • The sensor's reconfigurability allows it to achieve two biosimilar photoresponse characteristics with slow and fast reactions to light stimuli.
  • The research is published in the journal Nano-Micro Letters, Vol. 18, Issue 1, 2025.
  • The citation for the research is: Ultrathin Gallium Nitride Quantum-Disk-in-Nanowire-Enabled Reconfigurable Bioinspired Sensor for High-Accuracy Human Action Recognition.
  • The research is conducted at the Institute of Materials Research and Engineering, A*STAR, Singapore.

Sources:

  • Ultrathin Gallium Nitride Quantum-Disk-in-Nanowire-Enabled Reconfigurable Bioinspired Sensor for High-Accuracy Human Action Recognition. Nano-Micro Letters, 2025;18(1):54.
  • Shanghai Jiao Tong Univ Press, Shanghai Jiao Tong Univ, 800 Dongchuan Rd, Shanghai, 200240, Peoples R China.
  • Institute of Materials Research and Engineering, Fusionopolis Way, #08-03, Agency for Science Technology and Research, Singapore, 138634, Singapore.