Patents by Inventor Zixiang Liang

Zixiang Liang has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Training method of underwater sea urchin image recognition model, and underwater sea urchin image recognition method and device
    Patent number: 12198421
    Abstract: A training method of underwater sea urchin image recognition model, and underwater sea urchin image recognition method and device are provided. The training method includes: constructing an underwater sea urchin image recognition model, acquiring an underwater sea urchin image dataset; performing MSRCR on the underwater sea urchin image dataset; processing the underwater sea urchin image dataset by a dark channel prior method; performing image fusion to obtain a fused image dataset; sharpening the fused image dataset to obtain a training image dataset; and training the underwater sea urchin image recognition model by using the training image dataset. By performing MSRCR and dark channel prior processing on the image, and then performing image fusion and sharpening based on a point sharpness weighting method, the quality of image is improved, and an accuracy of subsequent sea urchin object detection is improved.
    Type: Grant
    Filed: February 8, 2023
    Date of Patent: January 14, 2025
    Assignee: Guangdong Ocean University
    Inventors: Xiaohong Peng, Zixiang Liang, Sidong Li, Liang Chen, Jun Zhang, Yifan Li, Xiaopeng Zhao
  • TRAINING METHOD OF UNDERWATER SEA URCHIN IMAGE RECOGNITION MODEL, AND UNDERWATER SEA URCHIN IMAGE RECOGNITION METHOD AND DEVICE
    Publication number: 20240020966
    Abstract: A training method of underwater sea urchin image recognition model, and underwater sea urchin image recognition method and device are provided. The training method includes: constructing an underwater sea urchin image recognition model, acquiring an underwater sea urchin image dataset; performing MSRCR on the underwater sea urchin image dataset; processing the underwater sea urchin image dataset by a dark channel prior method; performing image fusion to obtain a fused image dataset; sharpening the fused image dataset to obtain a training image dataset; and training the underwater sea urchin image recognition model by using the training image dataset. By performing MSRCR and dark channel prior processing on the image, and then performing image fusion and sharpening based on a point sharpness weighting method, the quality of image is improved, and an accuracy of subsequent sea urchin object detection is improved.
    Type: Application
    Filed: February 8, 2023
    Publication date: January 18, 2024
    Inventors: Xiaohong Peng, Zixiang Liang, Sidong Li, Liang Chen, Jun Zhang, Yifan Li, Xiaopeng Zhao