Automated Bale Mapping Using Machine Learning and Photogrammetry

Abstract

An automatic method of obtaining geographic coordinates of bales using monovision un-crewed aerial vehicle imagery was developed utilizing a data set of 300 images with a 20-megapixel resolution containing a total of 783 labeled bales of corn stover and soybean stubble. The relative performance of image processing with Otsu’s segmentation, you only look once version three (YOLOv3), and region-based convolutional neural networks was assessed. As a result, the best option in terms of accuracy and speed was determined to be YOLOv3, with 80% precision, 99% recall, 89% F1 score, 97% mean average precision, and a 0.38 s inference time. Next, the impact of using lower-cost cameras was evaluated by reducing image quality to one megapixel. The lower-resolution images resulted in decreased performance, with 79% precision, 97% recall, 88% F1 score, 96% mean average precision, and 0.40 s inference time. Finally, the output of the YOLOv3 trained model, density-based spatial clustering, photogrammetry, and map projection were utilized to predict the geocoordinates of the bales with a root mean squared error of 2.41 m.

Remote Sensing Object Detection in the Deep Learning Era—A Review

“…Agricultural applications: Other than typical crop-yield estimation applications [194,201], object detection has been used to optimize the agriculture industry. For example, at a higher resolution, remote sensing-based object detection was used for bale counting to automate and optimize the inventory estimation [202,203] and space and storage utilization. Their application automates the inventory process of bales, providing accurate counts and reducing manual labor.…”

An Intelligent Hierarchical Cyber-Physical System for Beach Waste Management: The BIOBLU Case Study

“…The geolocalization step aims at identifying the georeferenced locations of the detected waste items to enable the UGV robot collect them. The bounding box pixel coordinates provided by the detection object list are thus transformed into geographic information coordinates (GPS latitude and longitude) exploiting the equations of coordinate systems and map projections [37], [38] also considering the image metadata and UAV telemetry.…”