Design and implementation of a dual-feature fusion target tracking detection system based on Python Productivity for ZYNQ

Jiayu Zhao, Chunguang Huang, Fei Gao, Hai ChengICML 2022 · DOI

Abstract

To address the drawbacks of high power consumption and slow running speed of image processing algorithms on traditional hardware platforms, we design and implement a target tracking detection system based on PYNQ (Python productivity for Zynq) heterogeneous processor platform through a hardware-software co-design approach. In the PS (Process System) part of the processor system of PYNQ, the Haar and Edgelet dual-feature fusion training AdaBoost classifier is completed. In the Programmable Logic (PL) part of PYNQ, the acceleration design of the inter-frame difference method IP core is completed, and the motion target tracking detection is completed. The system was tested in terms of real-time performance and power consumption, and the results showed that the performance was better than that of traditional hardware platforms.
BibTeX
@inproceedings{zhao2022design,
  author = {Jiayu Zhao and Chunguang Huang and Fei Gao and Hai Cheng},
  title = {{Design and implementation of a dual-feature fusion target tracking detection system based on Python Productivity for ZYNQ}},
  booktitle = {International Conference on Machine Learning},
  year = {2022},
  doi = {10.1109/mlise57402.2022.00049},
}

Loading related content...

Figure