Design of a Reconfigurable Crop Scouting Vehicle for Row Crop Navigation: A Proof-of-Concept Study.

Pest infestation causes significant crop damage during crop production, which reduces the crop yield in terms of quality and quantity. Accurate, precise, and timely information on pest infestation is a crucial aspect of integrated pest management practices. The current manual scouting methods are time-consuming and laborious, particularly for large fields. Therefore, a fleet of scouting vehicles is proposed to monitor and collect crop information at the sub-canopy level. These vehicles would traverse large fields and collect real-time information on pest type, concentration, and infestation level. In addition to this, the developed vehicle platform would assist in collecting information on soil moisture, nutrient deficiency, and disease severity during crop growth stages. This study established a proof-of-concept of a crop scouting vehicle that can navigate through the row crops. A reconfigurable ground vehicle (RGV) was designed and fabricated. The developed prototype was tested in the laboratory and an actual field environment. Moreover, the concept of corn row detection was established by utilizing an array of low-cost ultrasonic sensors. The RGV was successful in navigating through the corn field. The RGV's reconfigurable characteristic provides the ability to move anywhere in the field without damaging the crops. This research shows the promise of using reconfigurable robots for row crop navigation for crop scouting and monitoring which could be modular and scalable, and can be mass-produced in quick time. A fleet of these RGVs would empower the farmers to make meaningful and timely decisions for their cropping system.

A New Dataset and Comparative Study for Aphid Cluster Detection and Segmentation in Sorghum Fields.

Aphid infestations are one of the primary causes of extensive damage to wheat and sorghum fields and are one of the most common vectors for plant viruses, resulting in significant agricultural yield losses. To address this problem, farmers often employ the inefficient use of harmful chemical pesticides that have negative health and environmental impacts. As a result, a large amount of pesticide is wasted on areas without significant pest infestation. This brings to attention the urgent need for an intelligent autonomous system that can locate and spray sufficiently large infestations selectively within the complex crop canopies. We have developed a large multi-scale dataset for aphid cluster detection and segmentation, collected from actual sorghum fields and meticulously annotated to include clusters of aphids. Our dataset comprises a total of 54,742 image patches, showcasing a variety of viewpoints, diverse lighting conditions, and multiple scales, highlighting its effectiveness for real-world applications. In this study, we trained and evaluated four real-time semantic segmentation models and three object detection models specifically for aphid cluster segmentation and detection. Considering the balance between accuracy and efficiency, Fast-SCNN delivered the most effective segmentation results, achieving 80.46% mean precision, 81.21% mean recall, and 91.66 frames per second (FPS). For object detection, RT-DETR exhibited the best overall performance with a 61.63% mean average precision (mAP), 92.6% mean recall, and 72.55 on an NVIDIA V100 GPU. Our experiments further indicate that aphid cluster segmentation is more suitable for assessing aphid infestations than using detection models.

Aphid cluster recognition and detection in the wild using deep learning models.

Aphid infestation poses a significant threat to crop production, rural communities, and global food security. While chemical pest control is crucial for maximizing yields, applying chemicals across entire fields is both environmentally unsustainable and costly. Hence, precise localization and management of aphids are essential for targeted pesticide application. The paper primarily focuses on using deep learning models for detecting aphid clusters. We propose a novel approach for estimating infection levels by detecting aphid clusters. To facilitate this research, we have captured a large-scale dataset from sorghum fields, manually selected 5447 images containing aphids, and annotated each individual aphid cluster within these images. To facilitate the use of machine learning models, we further process the images by cropping them into patches, resulting in a labeled dataset comprising 151,380 image patches. Then, we implemented and compared the performance of four state-of-the-art object detection models (VFNet, GFLV2, PAA, and ATSS) on the aphid dataset. Extensive experimental results show that all models yield stable similar performance in terms of average precision and recall. We then propose to merge close neighboring clusters and remove tiny clusters caused by cropping, and the performance is further boosted by around 17%. The study demonstrates the feasibility of automatically detecting and managing insects using machine learning models. The labeled dataset will be made openly available to the research community.

Dynamic Label Assignment for Object Detection by Combining Predicted IoUs and Anchor IoUs.

Label assignment plays a significant role in modern object detection models. Detection models may yield totally different performances with different label assignment strategies. For anchor-based detection models, the IoU (Intersection over Union) threshold between the anchors and their corresponding ground truth bounding boxes is the key element since the positive samples and negative samples are divided by the IoU threshold. Early object detectors simply utilize the fixed threshold for all training samples, while recent detection algorithms focus on adaptive thresholds based on the distribution of the IoUs to the ground truth boxes. In this paper, we introduce a simple while effective approach to perform label assignment dynamically based on the training status with predictions. By introducing the predictions in label assignment, more high-quality samples with higher IoUs to the ground truth objects are selected as the positive samples, which could reduce the discrepancy between the classification scores and the IoU scores, and generate more high-quality boundary boxes. Our approach shows improvements in the performance of the detection models with the adaptive label assignment algorithm and lower bounding box losses for those positive samples, indicating more samples with higher-quality predicted boxes are selected as positives.

Early-season plant-to-plant spatial uniformity can affect soybean yields.

Increased soybean (Glycine max L. Merril) seed costs have motivated interest in reduced seeding rates to improve profitability while maintaining or increasing yield. However, little is known about the effect of early-season plant-to-plant spatial uniformity on the yield of modern soybean varieties planted at reduced seeding rates. The objectives of this study were to (i) investigate traditional and devise new metrics for characterizing early-season plant-to-plant spatial uniformity, (ii) identify the best metrics correlating plant-to-plant spatial uniformity and soybean yield, and (iii) evaluate those metrics at different seeding rate (and achieved plant density) levels and yield environments. Soybean trials planted in 2019 and 2020 compared seeding rates of 160, 215, 270, and 321 thousand seeds ha-1 planted with two different planters, Max Emerge and Exact Emerge, in rainfed and irrigated conditions in the United States (US). In addition, trials comparing seeding rates of 100, 230, 360, and 550 thousand seeds ha-1 were conducted in Argentina (Arg) in 2019 and 2020. Achieved plant density, grain yield, and early-season plant-to-plant spacing (and calculated metrics) were measured in all trials. All site-years were separated into low- (2.7 Mg ha-1), medium- (3 Mg ha-1), and high- (4.3 Mg ha-1) yielding environments, and the tested seeding rates were separated into low (< 200 seeds m-2), medium (200-300 seeds m-2), and high (> 300 seeds m-2) levels. Out of the 13 metrics of spatial uniformity, standard deviation (sd) of spacing and of achieved versus targeted evenness index (herein termed as ATEI, observed to theoretical ratio of plant spacing) showed the greatest correlation with soybean yield in US trials (R2 = 0.26 and 0.32, respectively). However, only the ATEI sd, with increases denoting less uniform spacing, exhibited a consistent relationship with yield in both US and Arg trials. The effect of spatial uniformity (ATEI sd) on soybean yield differed by yield environment. Increases in ATEI sd (values > 1) negatively impacted soybean yields in both low- and medium-yield environments, and in achieved plant densities below 200 thousand plants ha-1. High-yielding environments were unaffected by variations in spatial uniformity and plant density levels. Our study provides new insights into the effect of early-season plant-to-plant spatial uniformity on soybean yields, as influenced by yield environments and reduced plant densities.