Accurate and fast implementation of soybean pod counting and localization from high-resolution image

Introduction

Soybean pod count is one of the crucial indicators of soybean yield. Nevertheless, due to the challenges associated with counting pods, such as crowded and uneven pod distribution, existing pod counting models prioritize accuracy over efficiency, which does not meet the requirements for lightweight and real-time tasks.

Methods

To address this goal, we have designed a deep convolutional network called PodNet. It employs a lightweight encoder and an efficient decoder that effectively decodes both shallow and deep information, alleviating the indirect interactions caused by information loss and degradation between non-adjacent levels.

Results

We utilized a high-resolution dataset of soybean pods from field harvesting to evaluate the model's generalization ability. Through experimental comparisons between manual counting and model yield estimation, we confirmed the effectiveness of the PodNet model. The experimental results indicate that PodNet achieves an R2 of 0.95 for the prediction of soybean pod quantities compared to ground truth, with only 2.48M parameters, which is an order of magnitude lower than the current SOTA model YOLO POD, and the FPS is much higher than YOLO POD.

Discussion

Compared to advanced computer vision methods, PodNet significantly enhances efficiency with almost no sacrifice in accuracy. Its lightweight architecture and high FPS make it suitable for real-time applications, providing a new solution for counting and locating dense objects.

Changing light promotes isoflavone biosynthesis in soybean pods and enhances their resistance to mildew infection

Mildew severely reduces soybean yield and quality, and pods are the first line of defence against pathogens. Maize-soybean intercropping (MSI) reduces mildew incidence on soybean pods; however, the mechanism remains unclear. Changing light (CL) from maize shading is the most important environmental feature in MSI. We hypothesized that CL affects isoflavone accumulation in soybean pods, affecting their disease resistance. In the present study, shading treatments were applied to soybean plants during different developmental stages according to various CL environments under MSI. Chlorophyll fluorescence imaging (CFI) and classical evaluation methods confirmed that CL, especially vegetative stage shading (VS), enhanced pod resistance to mildew. Further metabolomic analyses and exogenous jasmonic acid (JA) and biosynthesis inhibitor experiments revealed the important relationship between JA and isoflavone biosynthesis, which had a synergistic effect on the enhanced resistance of CL-treated pods to mildew. VS promoted the biosynthesis and accumulation of constitutive isoflavones upstream of the isoflavone pathway, such as aglycones and glycosides, in soybean pods. When mildew infects pods, endogenous JA signalling stimulated the biosynthesis of downstream inducible malonyl isoflavone (MIF) and glyceollin to improve pod resistance.

Pod Mildew on Soybeans Can Mitigate the Damage to the Seed Arising from Field Mold at Harvest Time

Seedpods are the outermost barrier of legume plants encountered by pests and pathogens, but research on this tissue, especially regarding their chemical constituents, is limited. In the present study, a mildew-index-model-based cluster analysis was used to evaluate and identify groups of soybean genotypes with different organ-specific resistance against field mold. The constituents of soybean pods, including proteins, carbohydrates, fatty acids, and isoflavones, were analyzed. Linear regression and correlation analyses were also conducted between these main pod constituents and the organ-specific mildew indexes of seed (MIS) and pod (MIP). With increases in the contents of infection constituents, such as proteins, carbohydrates, and fatty acids, the MIP increased and the MIS decreased. The MIS decreased with increases in the contents of glycitein (GLE)-type isoflavonoids, which act as antibiotic constituents. Although the infection constituents in the soybean pods caused pod mildew, they also helped mitigate the corresponding seed mildew to a certain extent.