Unmanned aerial vehicle (UAV) has become a popular technology, and it has promoted the development of many applications in different areas. In the context of precision agriculture, UAV’s images enable us to identify the location of planting rows in order to plan and to estimate the crop production and the number of plants, as well as early identification and correction of failures in sowing. As these applications deal with low- or medium-altitude imagery, new image processing techniques are necessary to process these images. This paper proposes an automatic segmentation approach that uses a genetic algorithm (GA) and Radon transform to detect sugarcane crop lines from images obtained using a UAV assembled with an RGB sensor.

无人机（UAV）已成为一种流行的技术，并促进了不同领域中许多应用程序的发展。在精准农业的背景下，UAV的图像使我们能够识别种植行的位置，以计划和估计作物产量和植物数量，以及及早发现和纠正播种失败的情况。由于这些应用程序处理低海拔或中等高度的图像，因此需要新的图像处理技术来处理这些图像。本文提出了一种自动分割方法，该方法使用遗传算法（GA）和Radon变换从使用RGB传感器组装的无人机获得的图像中检测甘蔗作物行。