范圣哲, 贡亮, 杨智宇, 等. 面向水稻穗上谷粒原位计数与遮挡还原的轻量级I2I深度学习方法[J]. 华南农业大学学报, 2023, 44(1): 74-83. DOI: 10.7671/j.issn.1001-411X.202202008
    引用本文: 范圣哲, 贡亮, 杨智宇, 等. 面向水稻穗上谷粒原位计数与遮挡还原的轻量级I2I深度学习方法[J]. 华南农业大学学报, 2023, 44(1): 74-83. DOI: 10.7671/j.issn.1001-411X.202202008
    FAN Shengzhe, GONG Liang, YANG Zhiyu, et al. A lightweight I2I deep learning method for on-panicle grain in-situ counting and occluded grains restoration[J]. Journal of South China Agricultural University, 2023, 44(1): 74-83. DOI: 10.7671/j.issn.1001-411X.202202008
    Citation: FAN Shengzhe, GONG Liang, YANG Zhiyu, et al. A lightweight I2I deep learning method for on-panicle grain in-situ counting and occluded grains restoration[J]. Journal of South China Agricultural University, 2023, 44(1): 74-83. DOI: 10.7671/j.issn.1001-411X.202202008

    面向水稻穗上谷粒原位计数与遮挡还原的轻量级I2I深度学习方法

    A lightweight I2I deep learning method for on-panicle grain in-situ counting and occluded grains restoration

    • 摘要:
      目的  为解决传统水稻考种机谷粒表型分析算法在功能和效率上的局限性,针对穗上谷粒原位计数和被遮挡谷粒几何特征还原设计一种基于深度学习的轻量级通用算法框架。
      方法  将穗上谷粒原位计数与被遮挡谷粒还原这2个复杂任务分别拆解为2个阶段,将其核心阶段建模为I2I问题。基于MobileNet V3设计1种能够解决I2I问题的轻量级网络架构,并针对2个任务的特点分别设计了数据集图像制作方法,选择合适的优化策略和超参数对其进行训练。训练结束后,使用TensorFlow Lite runtime解释器将模型部署在考种机的树莓派4B开发板上,并进行测试。
      结果  该算法在穗上谷粒计数任务中具有良好的准确性、快速性,且具有一定的泛化性能。在被遮挡谷粒的形状还原任务中,该算法所还原的谷粒图像在面积、周长、长度、宽度和颜色分数评价指标中准确率均达到97%以上。
      结论  该算法能够有效地完成穗上谷粒计数和被遮挡谷粒的还原任务,且具有轻量级的优点。

       

      Abstract:
      Objective  To address the functional and efficiency limitations of the conventional grain phenotype analysis algorithm of seed analyzers, a deep learning based lightweight general algorithmic framework was designed for two tasks: In-situ counting of on-panicle grains and restoration of occluded grains.
      Method  Two complex tasks of on-panicle grains in-situ counting and restoration of occluded grains were decomposed into two stages, and their core stages were modeled as I2I problems. A lightweight network architecture capable of solving the I2I problem was designed based on MobileNet V3, and the data set generation method was designed according to the characteristics of these two tasks. Then the network was trained with appropriate optimization strategies and hyperparameters. After training, the model was deployed and tested with TensorFlow Lite runtime on Raspberry Pi 4B development board.
      Result  The algorithm had good accuracy, rapidity and some generalizable performance in the task of on-panicle grain counting. In the task of occluded grains shape restoration, the evaluation accuracy of the restored images in the metrics of area, perimeter, length, width and color score were all over 97%.
      Conclusion  The algorithm proposed in this paper can complete the task of on-panicle grain counting and occluded grains restoration effectively, and also has the advantage of being lightweight.

       

    /

    返回文章
    返回