AgriEngineering, Volume 2, Issue 3 (September 2020) – 8 articles

Weed management is one of the most important aspects of crop productivity; knowing the amount and the locations of weeds has been a problem that experts have faced for several decades. This paper presents three methods for weed estimation based on deep learning image processing in lettuce crops, and we compared them to visual estimations by experts. One method is based on support vector machines (SVM) using histograms of oriented gradients (HOG) as feature descriptor. The second method was based in YOLOV3 (you only look once V3), taking advantage of its robust architecture for object detection, and the third one was based on Mask R-CNN (region based convolutional neural network) in order to get an instance segmentation for each individual. These methods were complemented with a NDVI index (normalized difference vegetation index) as a background subtractor for removing non photosynthetic objects. According to chosen metrics, the machine and deep learning methods had F1-scores of 88%, 94%, and 94% respectively, regarding to crop detection. Subsequently, detected crops were turned into a binary mask and mixed with the NDVI background subtractor in order to detect weed in an indirect way. Once the weed image was obtained, the coverage percentage of weed was calculated by classical image processing methods. Finally, these performances were compared with the estimations of a set from weed experts through a Bland–Altman plot, intraclass correlation coefficients (ICCs) and Dunn’s test to obtain statistical measurements between every estimation (machine-human); we found that these methods improve accuracy on weed coverage estimation and minimize subjectivity in human-estimated data. Full article

Carqueja (Baccharis crispa Spreng.) has been primarily used as a medicinal plant around the world. Commercially, the essential oil content of carqueja leaves is the most valuable crop productivity variable. We evaluated the effect of irrigation management in different growing seasons on the essential oil content of carqueja leaves using gas chromatography coupled with mass spectrometry. The experiment was conducted in a greenhouse located in Southern Brazil, where the crop was cultivated for two years in different growing seasons under six irrigation regimes: 25%, 50%, 75%, 100%, 125%, and 150% of the reference crop evapotranspiration (T25, T50, T75, T100, T125, and T150, respectively). A seasonal pattern was observed in the number of metabolites of sesquiterpenes and phenolics in the essential oil extracted from the biomass; this outcome was correlated with irrigation regimes and air temperature. Principal component and hierarchical cluster analyses were used to discriminate the influence of abiotic conditions on secondary metabolite profiles. Spathulenol was the most abundant compound in the essential oils (95.43%) collected during the summer (December–March) season during the third harvest (H3) at T150. The essential oil content was 8.84% ± 0.05% and 10.52% ± 0.10% in summer and winter (June–September), respectively, with T100 at 45 and 46 days after planting. Full article

Model simulations for bioenergy harvest planning need to utilize equipment-capacity relationships for equipment operating under the high-yield conditions typical of a biomass crop. These performance assumptions have a direct bearing on the estimates of machine capacity, the number of machines required, and, therefore, the cost to fulfill the biorefinery plant demands for a given harvest window. Typically, two major issues in these models have been poorly understood: the available time required to complete the harvest operation (often called probability of workdays) and the capacity of the harvest equipment as impacted by yield. Simulations use annual yield estimates, which incorporate weather events, to demonstrate year-to-year effects. Some simulations also incorporate potential yield increases from genetically modified energy crops. There are limited field performance data for most current forage equipment used for harvesting high-yield biomass crops. Analysis shows that the impact of wrap/eject time for round balers resulted in a 50% reduction in achieved throughput capacity (Mg/h). After the maximum throughput is reached, the cost of the round bale operation (3.23 USD/Mg) is double that of the large-square baler (1.63 USD/Mg). The round baler achieved throughput capacity is 50% less (32.7 Mg/h compared to 71.0 Mg/h) than the large-square baler. Full article

Plant stress is one of major issues that cause significant economic loss for growers. The labor-intensive conventional methods for identifying the stressed plants constrain their applications. To address this issue, rapid methods are in urgent needs. Developments of advanced sensing and machine learning techniques trigger revolutions for precision agriculture based on deep learning and big data. In this paper, we reviewed the latest deep learning approaches pertinent to the image analysis of crop stress diagnosis. We compiled the current sensor tools and deep learning principles involved in plant stress phenotyping. In addition, we reviewed a variety of deep learning applications/functions with plant stress imaging, including classification, object detection, and segmentation, of which are closely intertwined. Furthermore, we summarized and discussed the current challenges and future development avenues in plant phenotyping. Full article

The problem of finding an optimal solution for the slurry application process is casted as a capacitated vehicle routing problem (CVRP) in which by considering the vehicle’s capacity, it is required to visit all the tracks only once to fully cover the field, as well as complying with a specified targeted application rate. A key objective in this study was to determine an optimized coverage plan in order to minimize the driving distance in the field, while at the same time allowing for varying the application rate. The coverage plan includes the optimal sequence of tracks with a specified application rate for each track. Two algorithms were developed for optimization and simulation of the slurry application cast as capacitated operations. In order to validate the proposed algorithms, a slurry application operation was recorded, and the results of the optimization algorithm were compared with the conventional non-optimized method. The comparison showed that applying the proposed new method reduces the non-working distance by 18.6% and the non-working time by 28.1%. Full article

The aims of this research were to evaluate the influence of intermittent microwave drying on the moisture diffusion and color qualities of organically and conventionally grown sweet red peppers and mathematically express drying kinetic data. Pepper samples of 150 g were dried at 150, 300 and 450 W using a microwave oven. Results showed that intermittent microwave drying at 450 W occurred mainly in the falling rate period, whereas drying at lower powers resulted in relatively longer constant rate periods for both peppers types. The Midilli model provided the best fit for all data. The moisture diffusivity (D_eff) values of organic and conventional samples ranged from 59.69 × 10⁻¹⁰ to 182.01 × 10⁻¹⁰ m²s⁻¹ and from 59.11 × 10⁻¹⁰ to 181.01 × 10⁻¹⁰ m²s⁻¹, respectively, and the difference was insignificant. The pre-exponential factor for the Arrhenius equation (D₀) and activation energy (E_a) values were almost identical for both product types. Overall, organic or conventional growing did not alter the structural features related to the heat transfer properties. Intermittent microwave drying at 150 and 300 W for organic peppers and 150 W for conventional peppers gave the highest ΔL*, Δa* and a*/b* values, producing the most bright and red pepper powders. Thus, these treatments can be used to produce higher color quality powders. Full article

The present study comparatively investigates the potential of waste plastic utilization as a feedstock for the production of liquid fuels to satisfy the rising liquid fuel demands of the transportation industry while simultaneously resolving the global plastic waste pollution challenge. For the first time, therefore, conceptual models simulating the production of transportation fuels of ethanol and gasoline from waste plastics via the technologies of thermo-syngas fermentation and hydrothermal liquefaction were assessed using classic technoeconomic assessment methods. The conceptual models were developed based on existing experimental data as obtained from the literature and simulated using ASPEN Plus as the preferred process simulation tool. This study demonstrated the technical viability of both conversion pathways with the hydrothermal liquefaction (HTL) of waste plastics for gasoline production shown to constitute a more economically preferable pathway. This was because the HTL of waste plastics presented a higher internal rate of return (IRR) value and a lower unit processing cost of 51.3% and USD 0.38 per kg compared to the thermo-syngas fermentation pathway that presented an IRR value and a unit processing cost value of 22.2% and USD 0.42 per kg, respectively. Payback periods of 5 years and 2 years were also determined as vital to recoup initial capital invested in the thermo-syngas fermentation project and the HTL project, respectively. Therefore, this study provides a basis for further work regarding waste plastic management strategies while offering a useful guide for policy makers in determining the most cost-effective way to utilize waste plastic and thus promote favorable environmental outcomes. Full article

The conversion of pig manure into organic fertilizer has become a research hotspot in agricultural engineering, and many types of pig manure processing machinery have been derived. The discrete element method (DEM) can be used in the research of pig manure processing machinery to study the interaction between pig manure and machinery, which makes the research more direct and accurate. In order to introduce the discrete element method into the research of pig manure processing machinery, a reliable parameter basis for discrete element simulation is necessary, taking the angle of repose (AoR) as the reference and based on the hertz-mindlin with JKR contact model and Plackett–Burman experiment design. Three parameters with significant influence on the AoR are screened out from nine parameters related to pig manure. By conducting Box–Behnken experiment design, the quadratic polynomial regression equation between the AoR and three significant parameters is established. According to the parameters predicted by the quadratic polynomial regression equation, the discrete element simulation of AoR is conducted. The simulation result of AoR (38.54°) is close to the experimental result (38.65°) with a relative error of 0.28%, indicating that the regression equation can predict the relevant parameters of pig manure according to the AoR. Full article