Physiological maturity of bananas is of vital importance in determination of their quality and marketability. This study assessed, with the use of a Vis/NIR hyperspectral imaging (400–1000 nm), the feasibility in differentiating six maturity levels (maturity level 2, 4, and 6 to 9) of green dwarf banana and characterizing their quality changes during maturation. Spectra were extracted from three zones (pedicel, middle and apex zone) of each banana finger, respectively. Based on spectra of each zone, maturity identification models with high accuracy (all over 91.53% in validation set) were established by partial least squares discrimination analysis (PLSDA) method with raw spectra. A further generic PLSDA model with an accuracy of 94.35% for validation was created by the three zones’ spectra pooled to omit the effect of spectra acquisition position. Additionally, a spectral interval was selected to simplify the generic PLSDA model, and an interval PLSDA model was built with an accuracy of 85.31% in the validation set. For characterizing some main quality parameters (soluble solid content, SSC; total acid content, TA; chlorophyll content and total chromatism, ΔE*) of banana, full-spectra partial least squares (PLS) models and interval PLS models were, respectively, developed to correlate those parameters with spectral data. In full-spectra PLS models, high coefficients of determination (R²) were 0.74 for SSC, 0.68 for TA, and fair of 0.42 as well as 0.44 for chlorophyll and ΔE*. The performance of interval PLS models was slightly inferior to that of the full-spectra PLS models. Results suggested that models for SSC and TA had an acceptable predictive ability (R² = 0.64 and 0.59); and models for chlorophyll and ΔE* (R² = 0.34 and 0.30) could just be used for sample screening. Visualization maps of those quality parameters were also created by applying the interval PLS models on each pixel of the hyperspectral image, the distribution of quality parameters in which were basically consistent with the actual measurement. This study proved that the hyperspectral imaging is a useful tool to assess the maturity level and quality of dwarf bananas.

In the traditional agricultural wireless sensor networks (WSNs), there is a large amount of redundant data and high latency on critical events (CEs) for data collection systems, which increases the time and energy consumption. In order to overcome these problems, an effective edge computing (EC) enabled data collection approach for CE in smart agriculture is proposed. First, the key features data types (KFDTs) are extracted from the historical dataset to keep the main information on CEs. Next, the KFDTs are selected as the collection data type of the software-defined wireless sensor network (SDWSN). Then, the event types are decided by searching the minimum average variance between the sensing data of active nodes and the average value of the key feature data obtained by EC. Furthermore, the sensing nodes are driven to sense the event-related data with a consideration of latency constraints by the SDWSN servers. A real-world testbed was set up in a smart greenhouse for experimental verification of the proposed approach. The results showed that the proposed approach could reduce the number of needed sensors, sensing time, collection data volume, communication time, and provide the low latency agricultural data collection system. Thus, the proposed approach can improve the efficiency of CE sensing in smart agriculture.

In smart agricultural systems, the macroinformation sensing by adopting a mobile robot with multiple types of sensors is a key step for sustainable development of agriculture. Also, in a region monitoring strategy that meets the real-scene requirements, optimal operation of mobile robots is necessary. In this paper, a cloud-assisted region monitoring strategy of mobile robots in a smart greenhouse is presented. First, a hybrid framework that contains a cloud, a wireless network, and mobile multisensor robots is deployed to monitor a wide-region greenhouse. Then, a novel strategy that contains two phases is designed to ensure valid region monitoring and meet the time constraints of a mobile sensing robot. In the first phase, candidate region monitoring points are selected using the improved virtual forces. In the second phase, a moving path for the mobile node is calculated based on Euclidean distance. Subsequently, the applicability of the proposed strategy is verified by the greenhouse test system. The verification results show that the proposed algorithm has better performance than the conventional methods. The results also demonstrate that, by applying the proposed algorithm, the number of monitoring points and time consumption can reduce, while the valid monitoring region area is enlarged.

White clover (Trifolium repens L.) and ryegrass (Lolium multiflorum L.) are widely used cover crops. This experiment investigated the potential of white clover (WC), ryegrass (RG), and fallow (FL) to inhibit the growth of weeds and the effect of their residue return to the field on subsequent crops in a cover crop-corn rotation system. Furthermore, we designed pot experiments to guide the scientific application of WC and RG. The results showed that the FL treatment had the highest mean weed biomass in two years (11.99 t ha⁻¹) and the RG treatment recorded the lowest mean weed biomass in two years (2.04 t ha⁻¹) as its early growth rate and aerial root cover. The combination of rotary tillage (RT) and WC recorded the highest total corn yield in two years (20.20 t ha⁻¹) and an increase of 2.84% in the two-year average biomass of weeds compared to RT-FL. Compared to RT-FL, RT-RG treatments inhibited weed invasion by an average of 73%, but corn yield was reduced by 3.25%. Straw and soil ratios above 6:100 for RG resulted in stunted corn growth, including a reduction in fresh weight and chlorophyll content, and impaired photosynthesis, but this effect was not evident in WC treatment. From an ecological point of view, this study recommends RT-RG as a pre-crop for corn to reduce herbicide use.

Neural networks are widely used in fruit sorting and have achieved some success. However, due to the limitations of storage space and power consumption, the storage and computing of a neural network model on embedded devices remain a massive challenge. Aiming at realizing a lightweight mango sorting model, the feature-extraction characteristics of the shallow and deep networks of the SqueezeNet model were analyzed by a visualization method, and then eight lightweight models were constructed by removing redundant layers or modifying the convolution kernel. It was found that the model designated Model 4 performed well after training and testing. The class activation mapping method was used to explain the basis of the classification decision, and the model was compared with ten classical classification models. The results showed that the calculation performance of the model was significantly improved without reducing accuracy. The parameter storage requirement is 0.87 MB, and the calculation amount is 181 MFLOPS, while the average classification accuracy can still be maintained at 95.64%. This model has a high-cost performance and can be widely used in embedded devices.

Astragalus sinicus L. (milk vetch), a versatile plant that has a soil-enriching effect as green manure, is widely planted in the temperate zone of China. In previous experiments, milk vetch incorporated into the soil as green manure showed potential for goosegrass control. However, “what exactly happens at the chemical level?” and “what are the compounds that are potentially responsible for the phytotoxic effects observed during those previous assays?” In a recent study, in vitro phytotoxicity bioassays and chemical analyses of milk vetch decomposition leachates were carried out to explore the relationship between the temporal phytotoxic effects and the dynamics of chemical composition. For that, milk vetch decomposition leachates with a decay time of 12 h, 9 days, 12 days, 15 days, and 18 days were analyzed for organic compounds by liquid chromatography. The main results were as follows: (1) three compounds with goosegrass suppression potential produced during the decomposed process, i.e., 4-ethylphenol, N-acrylimorpholine, and allyl isothiocyanate. 2-Hydroxyethyl acrylate was present in the 12-h decomposition leachates but was at its highest concentration of 127.1 µg ml⁻¹ at 15 days. (2) The cultures were configured according to the four concentrations of goosegrass-resistant active substances measured in the 15-day decomposition leachate and, as with the 15-day decomposition leachate, the mixture cultures inhibited 100% of goosegrass germination at the high concentrations (≥ 30%), which suggests that these substances have goosegrass suppression potential. (3) The high total phenolic content (302.8–532.3 mg L⁻¹), the total flavonoid content (8.4–72.1 mg L⁻¹), and the reducing activity of the decomposition leachates for different decay times may explain why the incorporation of milk vetch into the soil did not lead to peroxidation of goosegrass in the previous study. (4) Finally, the changes in acid fraction and total content (1.9–4.2 mg ml⁻¹) for different decay times explain the variations in pH of the decomposition leachates, which, when discussed in conjunction with previous studies, may lead to changes in soil nutrient effectiveness and consequently affect crop growth. This study can provide a reference for green weed control research.