Wheat yield is influenced by the number of ears per unit area, and manual counting has traditionally been used to estimate wheat yield. To realize rapid and accurate wheat ear counting, K-means clustering was used for the automatic segmentation of wheat ear images captured by hand-held devices. The segmented data set was constructed by creating four categories of image labels: non-wheat ear, one wheat ear, two wheat ears, and three wheat ears, which was then was sent into the convolution neural network (CNN) model for training and testing to reduce the complexity of the model. The recognition accuracy of non-wheat, one wheat, two wheat ears, and three wheat ears were 99.8, 97.5, 98.07, and 98.5%, respectively. The model R2 reached 0.96, the root mean square error (RMSE) was 10.84 ears, the macro F1-score and micro F1-score both achieved 98.47%, and the best performance was observed during late grain-filling stage (R2 = 0.99, RMSE = 3.24 ears). The model could also be applied to the UAV platform (R2 = 0.97, RMSE = 9.47 ears). The classification of segmented images as opposed to target recognition not only reduces the workload of manual annotation but also improves significantly the efficiency and accuracy of wheat ear counting, thus meeting the requirements of wheat yield estimation in the field environment.

中文翻译：

---

使用 K 均值聚类分割和卷积神经网络进行麦穗计数。


小麦产量受单位面积穗数的影响，传统上采用人工计数来估算小麦产量。为了实现快速、准确的麦穗计数，采用K均值聚类对手持设备拍摄的麦穗图像进行自动分割。通过创建四类图像标签来构建分割数据集：非麦穗、一麦穗、二麦穗和三麦穗，然后将其送入卷积神经网络（CNN）模型进行训练和测试以降低模型的复杂度。非麦、一麦、二麦穗、三麦穗的识别准确率分别为99.8%、97.5%、98.07%、98.5%。模型R2达到0.96，均方根误差（RMSE）为10.84穗，宏观F1值和微观F1值均达到98.47%，且在灌浆后期表现最好（R2=0.99， RMSE = 3.24 耳）。该模型也可以应用于无人机平台（R2 = 0.97，RMSE = 9.47 耳）。与目标识别相比，分割图像的分类不仅减少了人工标注的工作量，而且显着提高了麦穗计数的效率和准确性，从而满足了田间环境下小麦估产的要求。