In recent years, precision agriculture has been researched to increase crop production with less inputs, as a promising means to meet the growing demand of agriculture products. Computer vision-based crop detection with unmanned aerial vehicle (UAV)-acquired images is a critical tool for precision agriculture. However, object detection using deep learning algorithms rely on a significant amount of manually prelabeled training datasets as ground truths. Field object detection, such as bales, is especially difficult because of (1) long-period image acquisitions under different illumination conditions and seasons; (2) limited existing prelabeled data; and (3) few pretrained models and research as references. This work increases the bale detection accuracy based on limited data collection and labeling, by building an innovative algorithms pipeline. First, an object detection model is trained using 243 images captured with good illimitation conditions in fall from the crop lands. In addition, domain adaptation (DA), a kind of transfer learning, is applied for synthesizing the training data under diverse environmental conditions with automatic labels. Finally, the object detection model is optimized with the synthesized datasets. The case study shows the proposed method improves the bale detecting performance, including the recall, mean average precision (mAP), and F measure (F1 score), from averages of 0.59, 0.7, and 0.7 (the object detection) to averages of 0.93, 0.94, and 0.89 (the object detection + DA), respectively. This approach could be easily scaled to many other crop field objects and will significantly contribute to precision agriculture.

This study investigates the efficacy of handheld Near-Infrared Spectroscopy (NIRS) devices for in-field estimation of forage quality using undried samples. The objective is to assess the precision and accuracy of multiple handheld NIRS instruments—NeoSpectra, TrinamiX, and AgroCares—when evaluating key forage quality metrics such as Crude Protein (CP), Neutral Detergent Fiber (aNDF), Acid Detergent Fiber (ADF), Acid Detergent Lignin (ADL), in vitro Total Digestibility (IVTD)and Neutral Detergent Fiber Digestibility (NDFD). Samples were collected from silage bunkers across 111 farms in New York State and scanned using different methods (static, moving, and turntable). The results demonstrate that dynamic scanning patterns (moving and turntable) enhance the predictive accuracy of the models compared to static scans. Fiber constituents (ADF, aNDF) and Crude Protein (CP) show higher robustness and minimal impact from water interference, maintaining similar $R^2$ values as dried samples. Conversely, IVTD, NDFD, and ADL are adversely affected by water content, resulting in lower $R^2$ values. This study underscores the importance of understanding the water effects on undried forage, as water‘s high absorption bands at 1400 and 1900 nm introduce significant spectral interference. Further investigation into the PLSR loading factors is necessary to mitigate these effects. The findings suggest that, while handheld NIRS devices hold promise for rapid, on-site forage quality assessment, careful consideration of scanning methodology is crucial for accurate prediction models. This research contributes valuable insights for optimizing the use of portable NIRS technology in forage analysis, enhancing feed utilization efficiency, and supporting sustainable dairy farming practices.

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.