Development of a super high-speed vision platform that can process 1024×1024 images in real time at 12500 fps.

Kohei Shimasaki, Qingyi Gu, Jun Chen, Tadayoshi Aoyama, Takeshi Takaki, Idaku Ishii, Koichi Tamura, Kenji TajimaThe Proceedings of JSME Annual Conference on Robotics and Mechatronics (Robomec) 2016 · DOI

Abstract

This paper describes our developing super high-speed vision platform that can simultaneously process 12-bit gray-level 1024×1024 images at 12500 fps by implementing parallel hardware logic on the large-size FPGAs of the platform. Several experimental results demonstrate that our platform can calculate 4096-level brightness-histogram, 4096-level edge-intensity-histogram, and the 0th and 1st moment features for centroid calculation, for 1024×1024 images at 12500 fps.
BibTeX
@inproceedings{shimasaki2016development,
  author = {Kohei Shimasaki and Qingyi Gu and Jun Chen and Tadayoshi Aoyama and Takeshi Takaki and Idaku Ishii and Koichi Tamura and Kenji Tajima},
  title = {{Development of a super high-speed vision platform that can process 1024×1024 images in real time at 12500 fps.}},
  booktitle = {The Proceedings of JSME Annual Conference on Robotics and Mechatronics (Robomec)},
  year = {2016},
  doi = {10.1299/jsmermd.2016.2a1-19b3},
}

Loading related content...

Figure