Abstract:The ability to automatically monitor agricultural fields is an important capability in precision farming, enabling steps towards more sustainable agriculture. Precise, high-resolution monitoring is a key prerequisite for targeted intervention and the selective application of agro-chemicals. The main goal of this paper is developing a novel crop/weed segmentation and mapping framework that processes multispectral images obtained from an unmanned aerial vehicle (UAV) using a deep neural network (DNN). Most studies on crop/weed semantic segmentation only consider single images for processing and classification. Images taken by UAVs often cover only a few hundred square meters with either color only or color and near-infrared (NIR) channels. Although a map can be generated by processing single segmented images incrementally, this requires additional complex information fusion techniques which struggle to handle high fidelity maps due to their computational costs and problems in ensuring global consistency. Moreover, computing a single large and accurate vegetation map (e.g., crop/weed) using a DNN is non-trivial due to difficulties arising from: (1) limited ground sample distances (GSDs) in high-altitude datasets, (2) sacrificed resolution resulting from downsampling high-fidelity images, and (3) multispectral image alignment. To address these issues, we adopt a stand sliding window approach that operates on only small portions of multispectral orthomosaic maps (tiles), which are channel-wise aligned and calibrated radiometrically across the entire map. We define the tile size to be the same as that of the DNN input to avoid resolution loss. Compared to our baseline model (i.e., SegNet with 3 channel RGB (red, green, and blue) inputs) yielding an area under the curve (AUC) of [background=0.607, crop=0.681, weed=0.576], our proposed model with 9 input channels achieves [0.839, 0.863, 0.782]. Additionally, we provide an extensive analysis of 20 trained models, both qualitatively and quantitatively, in order to evaluate the effects of varying input channels and tunable network hyperparameters. Furthermore, we release a large sugar beet/weed aerial dataset with expertly guided annotations for further research in the fields of remote sensing, precision agriculture, and agricultural robotics.

A field assessment of a potential method for weed and crop mapping on the basis of crop planting geometry

Automated Crop and Weed Monitoring in Widely Spaced Cereals

Digital camera based measurement of crop cover for wheat yield prediction

Study On Field Weed Recognition In Real-Time

Computer-Vision-Based Weed Identification of Images Acquired by 3ccd Camera

Fusion of Pixel and Object-Based Features for Weed Mapping Using Unmanned Aerial Vehicle Imagery

A Field Weed Density Evaluation Method Based on UAV Imaging and Modified U-Net

Automated crop plant detection based on the fusion of color and depth images for robotic weed control

An Approach to the Use of Depth Cameras for Weed Volume Estimation

Machine learning approaches to automate weed detection by UAV based sensors

Detection and Analysis of Weed Impact on Sugar Beet Crop Using Drone Imagery

Area measurement for field crops based on UAV platform and image analysis

Automated crop plant counting from very high-resolution aerial imagery

Crop recognition under weedy conditions based on 3D imaging for robotic weed control

Site-specific weed management in corn using UAS imagery analysis and computer vision techniques

Estimation of Wheat Plant Density at Early Stages Using High Resolution Imagery

Weed Detection Based on the Fusion of Multiple Image Processing Algorithms

Weed Density and Distribution Estimation for Precision Agriculture Using Semi-Supervised Learning

A New Method for Crop Row Detection Using Unmanned Aerial Vehicle Images

WeedMap: A Large-Scale Semantic Weed Mapping Framework Using Aerial Multispectral Imaging and Deep Neural Network for Precision Farming

Evaluation of Two Deep Learning-Based Approaches for Detecting Weeds Growing in Cabbage