高级检索
    李昊玥, 陈桂芬, 裴傲. 基于改进Mask R-CNN的奶牛个体识别方法研究[J]. 华南农业大学学报, 2020, 41(6): 161-168. DOI: 10.7671/j.issn.1001-411X.202003030
    引用本文: 李昊玥, 陈桂芬, 裴傲. 基于改进Mask R-CNN的奶牛个体识别方法研究[J]. 华南农业大学学报, 2020, 41(6): 161-168. DOI: 10.7671/j.issn.1001-411X.202003030
    shu
    LI Haoyue, CHEN Guifen, PEI Ao. Research on individual recognition of dairy cows based on improved Mask R-CNN[J]. Journal of South China Agricultural University, 2020, 41(6): 161-168. DOI: 10.7671/j.issn.1001-411X.202003030
    Citation: LI Haoyue, CHEN Guifen, PEI Ao. Research on individual recognition of dairy cows based on improved Mask R-CNN[J]. Journal of South China Agricultural University, 2020, 41(6): 161-168. DOI: 10.7671/j.issn.1001-411X.202003030
    shu

    基于改进Mask R-CNN的奶牛个体识别方法研究

    Research on individual recognition of dairy cows based on improved Mask R-CNN

      摘要
    • 摘要:
      目的  针对传统奶牛养殖中采用人工识别奶牛个体的方法效率低且主观性强的问题，提出一种基于改进Mask R-CNN的奶牛个体识别方法。
      方法  该方法对Mask R-CNN中的特征提取网络结构进行优化，采用嵌入SE block的ResNet-50网络作为Backbone，通过加权策略对图像通道进行筛选以提高特征利用率；针对实例分割时目标边缘定位不准确的问题，引入IoU boundary loss构建新的Mask损失函数，以提高边界检测的精度；对3000张奶牛图像进行训练、验证和测试。
      结果  改进Mask R-CNN模型的精度均值(AP)达100%，IoUMask达91.34%；与原始Mask R-CNN模型相比，AP提高了3.28%，IoUMask提高了5.92%。
      结论  本文所提方法具备良好的目标检测能力，可为复杂农场环境下的奶牛个体精准识别提供参考。

       

      Abstract:
      Objective  To propose an individual cow recognition method based on the improved Mask R-CNN algorithm, and solve the problem of low efficiency and strong subjectivity of artificially identifying individual cows in traditional dairy farming.
      Method  This method optimizes the feature extraction network structure in Mask R-CNN, adopts ResNet-50 network embedded in SE block as backbone, and selects image channels by weighting strategy to improve feature utilization. For the problem of inaccurate target edge positioning during instance segmentation, a boundary weighted loss function is added to construct a new Mask loss function to improve the accuracy of boundary detection. A total of 3000 cow images are trained, validated and tested.
      Result  The improved Mask R-CNN model had an average precision (AP) of 100% and IoUMask of 91.34%. Compared with the original Mask R-CNN model, AP increased by 3.28% and IoUMask increased by 5.92%.
      Conclusion  The proposed method has strong segmentation accuracy and robustness, and can provide a reference for accurate recognition of cow images under complex farming environment.

       

      • HTML全文
      • 参考文献(21)
      • 相关文章
      • 施引文献
    • 资源附件(0)

    /

    返回文章
    返回