In recent years, ground-penetrating radar (GPR) has become increasingly important as a nondestructive way to explore plant roots. Automatic recognition and localization of root objects from GPR images present a significant challenge. GPR images for the root system contain complicated hyperbolic signals that appear deformation depending on root size, orientation, aggregation degree, and soil background. This article presents a new deep learning approach, You Only Look Once v4 (YOLOv4)-hyperbola, which provides fully automatic recognition and localization of root objects from GPR images. YOLOv4-hyperbola improves the YOLOv4 architecture by introducing the keypoints detection branch in order to accurately locate roots while identifying them. The YOLOv4-hyperbola model was trained by combining field datasets and simulated datasets to simultaneously identify and locate hyperbolic features representing potential root objects across GPR images and evaluated on datasets of root detection from two experiments in the field. Compared with the randomized Hough transform (RHT) method, the proposed approach demonstrated higher accuracy and efficiency in root object detection on GPR images. YOLOv4-hyperbola was able to accurately recognize and locate abnormal hyperbolic signals caused by the complexity of the root system in nature. The validation on the two independent datasets showed that the proposed approach had good generalization and great application potential for real-time detection and location of roots over large areas in the field.

中文翻译：

---

通过基于 YOLOv4 的深度学习方法增强 GPR 图像中的自动根识别和定位

近年来，探地雷达（GPR）作为一种无损探测植物根系的方法变得越来越重要。从 GPR 图像中自动识别和定位根对象是一项重大挑战。根系的 GPR 图像包含复杂的双曲线信号，这些信号会根据根系大小、方向、聚集程度和土壤背景而出现变形。本文介绍了一种新的深度学习方法，You Only Look Once v4 (YOLOv4)-hyperbola，它提供了 GPR 图像中根对象的全自动识别和定位。YOLOv4-hyperbola 通过引入关键点检测分支改进了 YOLOv4 架构，以便在识别它们的同时准确定位根。YOLOv4-双曲线模型通过结合现场数据集和模拟数据集进行训练，以同时识别和定位代表 GPR 图像中潜在根对象的双曲线特征，并在现场两个实验的根检测数据集上进行评估。与随机霍夫变换 (RHT) 方法相比，该方法在 GPR 图像上的根对象检测中表现出更高的准确性和效率。YOLOv4-hyperbola能够准确识别和定位自然界根系复杂性引起的异常双曲线信号。对两个独立数据集的验证表明，所提出的方法具有良好的泛化性和巨大的应用潜力，可用于现场大面积根系的实时检测和定位。