Implementation of Dairy Cows Individual Identification Based on ZYNQ SoC

2019 
Individual identification technology of dairy cows plays an important role in modern automated farming, which facilitates the management of various behaviors of dairy cows and improves the cow's health protection. In this paper, the ZYNQ-7000 hardware platform is used to preprocess the real-time cow frame image obtained by the camera, and then compared with the cow images in the database, so as to determine the identity information of the cow in the current frame picture and complete the task of individual identification of dairy cows. After the camera acquires the cow picture, the image preprocessing is carried out in the programmable logic part. Image pre-processing IP cores are designed in Vivado HLS, including median filtering IP core and adaptive binary IP core. Additionally, the image comparison algorithm is completed in Vivado SDK, and the similarity comparison is performed between the SD card template image and the current frame picture in the DDR. Finally, the recognition results are shown in two ways: one is to output the corresponding cow number in the database through the serial port on the experiment host, and the other is to simulate the output result by turning on and off the LED on the development board.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    3
    References
    0
    Citations
    NaN
    KQI
    []