Centro de Investigaciones en Óptica A.C., Loma del Bosque 115, Leon 37150, Guanajuato, Mexico.
Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias-Campo Experimental Pabellón, Pabellon de Arteaga 20671, Aguascalientes, Mexico.
Sensors (Basel). 2022 Apr 14;22(8):3021. doi: 10.3390/s22083021.
Crop and weed discrimination in natural field environments is still challenging for implementing automatic agricultural practices, such as weed control. Some weed control methods have been proposed. However, these methods are still restricted as they are implemented under controlled conditions. The development of a sound weed control system begins by recognizing the crop and the different weed plants presented in the field. In this work, a classification approach of L. (Crop), narrow-leaf weeds (NLW), and broadleaf weeds (BLW) from multi-plant images are presented. Moreover, a large image dataset was generated. Images were captured in natural field conditions, in different locations, and growing stages of the plants. The extraction of regions of interest (ROI) is carried out employing connected component analysis (CCA), whereas the classification of ROIs is based on Convolutional Neural Networks (CNN) and compared with a shallow learning approach. To measure the classification performance of both methods, accuracy, precision, recall, and F1-score metrics were used. The best alternative for the weed classification task at early stages of growth and in natural corn field environments was the CNN-based approach, as indicated by the 97% accuracy value obtained.
在自然农田环境中进行作物和杂草识别对于实施自动农业实践（如杂草控制）仍然具有挑战性。已经提出了一些杂草控制方法。然而，这些方法仍然受到限制，因为它们是在受控条件下实施的。健全的杂草控制系统的开发始于识别田间的作物和不同的杂草植物。在这项工作中，提出了一种从多植物图像中分类 L.（作物）、窄叶杂草（NLW）和宽叶杂草（BLW）的方法。此外，还生成了一个大型图像数据集。图像是在自然田间条件下、不同地点和植物生长阶段拍摄的。感兴趣区域（ROI）的提取是通过连通分量分析（CCA）来完成的，而 ROI 的分类则基于卷积神经网络（CNN），并与浅层学习方法进行比较。为了衡量这两种方法的分类性能，使用了准确性、精度、召回率和 F1 分数等指标。对于生长早期和自然玉米田环境中的杂草分类任务，基于 CNN 的方法是最佳选择，因为获得了 97%的准确率值。
