Grand Challenges of Advanced Technologies in Sustainable Agriculture 4.0: Future Farming, Harvesting and Preservation

Recently, the use of small UAVs for monitoring agricultural land areas has been increasingly used by agricultural producers in order to improve crop yields. However, correctly interpreting the collected imagery data is still a challenging task. In this study, an automated pipeline for monitoring C. Annuum crops based on a deep learning model is implemented. The system is capable of performing inferences on the health status of individual plants, and to determine their locations and shapes in a georeferenced orthomosaic. Accuracy achieved on the classification task was 94.5. AP values among classes were in the range of $[63,100]$ for plant location boxes, and in $[40,80]$ for foliar area predictions. The methodology requires only RGB images, and so, it can be replicated for the monitoring of other types of crops by only employing consumer-grade UAVs. A comparison with random forest and large-scale mean shift segmentation methods which use predetermined features is presented. NDVI results obtained with multispectral equipment are also included. Full article

To solve the low accuracy of image feature matching in horticultural robot visual navigation, an innovative and effective image feature matching algorithm was proposed combining the improved Oriented FAST and Rotated BRIEF (ORB) and Lucas–Kanade (LK) optical flow algorithm. First, image feature points were extracted according to the adaptive threshold calculated using the Michelson contrast. Then, the extracted feature points were uniformed by the quadtree structure, which can reduce the calculated amount of feature matching, and the uniform ORB feature points were roughly matched to estimate the position of the feature points in the matched image using the improved LK optical flow. Finally, the Hamming distance between rough matching points was calculated for precise matching. Feature extraction and matching experiments were performed in four typical scenes: normal light, low light, high texture, and low texture. Compared with the traditional algorithm, the uniformity and accuracy of the feature points extracted by the proposed algorithm were enhanced by 0.22 and 50.47%, respectively. Meanwhile, the results revealed that the matching accuracy of the proposed algorithm increased by 14.59%, whereas the matching time and total time decreased by 39.18% and 44.79%, respectively. The proposed algorithm shows great potential for application in the visual simultaneous localization and mapping (V-SLAM) of horticultural robots to achieve higher accuracy of real-time positioning and map construction. Full article

This study examines the means of reducing the damage to the branches of Camellia oleifera in the process of vibration picking and solving the problems of low equipment-development efficiency and slow product renewal caused by using traditional test methods to determine vibration picking parameters. In this study, the optimal vibration parameters were determined by using the self-response (branch acceleration and strain) law of the Camellia oleifera tree, and finite element analysis and experiments are used to solve this problem. The 3D model of Camellia oleifera was built by Solidworks. The natural frequencies of Camellia oleifera were analyzed by modal analysis, the vibration frequency and amplitude were determined by harmonic response analysis, and transient analysis was used to compare with the test results. The results show that the optimal vibration frequency range of Camellia oleifera is 4~10 Hz, and the average correlation coefficient between the maximum synthetic acceleration and the simulated value is 0.85, which shows that the model can reliably predict the vibration response. At the same time, the best vibration parameters were determined to be 9 Hz, 60 mm and 10 s. Under these parameters, the abscission rate of the Camellia oleifera fruit was 90%, and the damage rate of the flower bud was 13%. The mechanized picking effect of Camellia oleifera fruit was good. This study can quickly determine the vibration picking parameters of Camellia oleifera fruit and effectively improve the development speed of vibration picking of Camellia oleifera fruit. Full article

With the rapid growth of population and the increasing demand for food worldwide, improving productivity in farming procedures is essential. Smart farming is a concept that emphasizes the use of modern technologies such as the Internet of Things (IoT) and artificial intelligence (AI) to enhance productivity in farming practices. In a smart farming scenario, large amounts of data are collected from diverse sources such as wireless sensor networks, network-connected weather stations, monitoring cameras, and smartphones. These data are valuable resources to be used in data-driven services and decision support systems (DSS) in farming applications. However, one of the major challenges with these large amounts of agriculture data is their immense diversity in terms of format and meaning. Moreover, the different services and technologies in a smart farming ecosystem have limited capability to work together due to the lack of standardized practices for data and system integration. These issues create a significant challenge in cooperative service provision, data and technology integration, and data-sharing practices. To address these issues, in this paper, we propose the platform approach, a design approach intended to guide building effective, reliable, and robust smart farming systems. The proposed platform approach considers six requirements for seamless integration, processing, and use of farm data. These requirements in a smart farming platform include interoperability, reliability, scalability, real-time data processing, end-to-end security and privacy, and standardized regulations and policies. A smart farming platform that considers these requirements leads to increased productivity, profitability, and performance of connected smart farms. In this paper, we aim at introducing the platform approach concept for smart farming and reviewing the requirements for this approach. Full article

Precision livestock farming (PLF) has spread to various countries worldwide since its inception in 2003, though it has yet to be widely adopted. Additionally, the advent of Industry 4.0 and the Internet of Things (IoT) have enabled a continued advancement and development of PLF. This modern technological approach to animal farming and production encompasses ethical, economic and logistical aspects. The aim of this review is to provide an overview of PLF and Industry 4.0, to identify current applications of this rather novel approach in different farming systems for food producing animals, and to present up to date knowledge on the subject. Current scientific literature regarding the spread and application of PLF and IoT shows how efficient farm animal management systems are destined to become. Everyday farming practices (feeding and production performance) coupled with continuous and real-time monitoring of animal parameters can have significant impacts on welfare and health assessment, which are current themes of public interest. In the context of feeding a rising global population, the agri-food industry and industry 4.0 technologies may represent key features for successful and sustainable development. Full article

Cymbidium is the most famous and widely distributed type of plant in the Orchidaceae family. It has extremely high ornamental and economic value. With the continuous development of the Cymbidium industry in recent years, it has become increasingly difficult to classify, identify, develop, and utilize orchids. In this study, a classification model GL-CNN based on a convolutional neural network was proposed to solve the problem of Cymbidium classification. First, the image set was expanded by four methods (mirror rotation, salt-and-pepper noise, image sharpening, and random angle flip), and then a cascade fusion strategy was used to fit the multiscale features obtained from the two branches. Comparing the performance of GL-CNN with other four classic models (AlexNet, ResNet50, GoogleNet, and VGG16), the results showed that GL-CNN achieves the highest classification prediction accuracy with a value of 94.13%. This model can effectively detect different species of Cymbidium and provide a reference for the identification of Cymbidium germplasm resources. Full article

The aim of this paper was to design a device for monitoring the work of irrigation technology (in our case, irrigation by sprinkler). Two devices for monitoring selected irrigation operating parameters for two hose reel irrigation machines were designed. During the monitored period of connection of the equipment to the sprinkler, 15 irrigation doses were carried out for both sprinklers. Irrigation operating characteristics working pressure, hose reel speed and selected weather conditions temperature and humidity were monitored. When evaluating the results, we proved the need to monitor the operation of the sprinkler not only by the coefficient of variation C_v, but also by introducing the coefficient of non-uniformity a. The results obtained indicate variability with respect to a particular irrigation dose and the applicable assessment method. The results were reviewed by one-way ANOVA analysis where observed coefficients and irrigation dose were considered as dependence factors. The results indicate a statistically significant impact of the applied quality coefficient of work and thus the impact of a particular device (p < 0.05, F_crit = 2.77). When evaluating the effect of the included irrigation dose, we also showed a statistically significant effect in both facilities (p < 0.05, F = 1.92). By checking the operation of the hose reel irrigation machine, we managed to successfully apply the proposed classifications, which also perform the function of fault prediction. The proposed facilities show that proper plant operation and timely response can help create more efficient and sustainable irrigation services, not only saving water but also reducing costs for the owner. Full article

According to the Food and Agriculture Organization of the United Nations, the world population will reach nine billion people in 2050, of which 75% will live in urban settlements. One of the biggest challenges will be meeting the demand for food, as farmland is being lost to climate change, water scarcity, soil pollution, among other factors. In this context, hydroponics, an agricultural method that dispenses with soil, provides a viable alternative to address this problem. Although hydroponics has proven its effectiveness on a large scale, there are still challenges in implementing this technique on a small scale, specifically in urban and suburban settings. Also, in rural communities, where the availability of suitable technologies is scarce. Paradigms such as the Internet of Things and Industry 4.0, promote Precision Agriculture on a small scale, allowing the control of variables such as pH, electrical conductivity, temperature, among others, resulting in higher production and resource savings. Full article

This paper presents an integrated aerial system for the identification of Amazonian Moriche palm (Mauritia flexuosa) in dense forests, by analyzing the UAV-captured RGB imagery using a Mask R-CNN deep learning approach. The model was trained with 478 labeled palms, using the transfer learning technique based on the well-known MS COCO framework©. Comprehensive in-field experiments were conducted in dense forests, yielding a precision identification of $98\%$. The proposed model is fully automatic and suitable for the identification and inventory of this species above 60 m, under complex climate and soil conditions. Full article

This study investigated the effect of preharvest 1-MCP treatment on maintaining ‘Rojo Brillante’ persimmon firmness. Early in the season, preharvest 1-MCP was applied 1, 7 and 10 days after ethephon treatment. The fruit firmness was evaluated during three different harvests and after the commercialization period of 3 d at 3 °C, plus 6 d at 20 °C. Late in the season, 1-MCP was applied 3 days before harvest in the fruit treated with gibberellic acid (GA) and then cold-stored for up to 60 days, plus a 6-day shelf life at 20 °C. The results showed that preharvest 1-MCP delayed the fruit softening induced by ethephon during the harvest period, and was the most effective treatment when performed 1 day after ethephon application. Therefore, preharvest 1-MCP extended the harvest period of ethephon-treated fruit. At the end of the season, preharvest 1-MCP had the same effect on maintaining the fruit firmness as the commercial postharvest application. Full article

Pesticide-free, 3-D, spacer fabrics (Plant Armor Generation (PA Gen) 1 and 2) were investigated for proof-of-concept as an insect barrier to protect plants and improve plant agronomics for organic farming. The time to 50% penetration (TP₅₀) for tobacco thrips, Frankliniella fusca (Hinds) adults in laboratory Petri dish bioassays was 30 and 175 min for PA Gen 1 and 2, respectively, and 12 min for the control (a commercially available, single layer-crop cover, Proteknet). PA Gen 2 was ≥90% resistant to penetration of unfed caterpillar neonates, Helicoverpa zea (Boddie), while the TP₅₀‘s for Gen 1 and Proteknet were 3.1 and 2.35 h, respectively. In small cage studies, PA Gen 2 covered potted cabbage plants were 100% resistant to penetration by these insects through 10 d after which the study was ended. In small field plot studies for 3 summer months, cabbage plants grew approximately twice as fast when covered versus not covered with Gen 1 and Gen 2 without the need for insecticides or herbicides. This was not observed for the control crop cover. Martindale abrasion tests demonstrated Gen 1 and 2 were at least 6- and 1.8-fold more durable than the control crop cover used. Data are also presented on percentage light, water, air, and water vapor penetration across each textile and operational temperatures and humidity for cabbage plants covered and uncovered in small field plots. Full article

Traditionally farmers monitor their crops employing their senses and experience. However, the human sensory system is inconsistent due to stress, health, and age. In this paper, we propose an agronomic application for monitoring the growth of Portos tomato seedlings using Kinect 2.0 to build a more accurate, cost-effective, and portable system. The proposed methodology classifies the tomato seedlings into four categories: The first corresponds to the seedling with normal growth at the time of germination; the second corresponds to germination that occurred days after; the third category entails exceedingly late germination where its growth will be outside of the estimated harvest time; the fourth category corresponds to seedlings that did not germinate. Typically, an expert performs this classification by analyzing ten percent of the randomly selected seedlings. In this work, we studied different methods of segmentation and classification where the Gaussian Mixture Model (GMM) and Decision Tree Classifier (DTC) showed the best performance in segmenting and classifying Portos tomato seedlings. Full article

The roots are a vital organ for plant growth and health. The opaque surrounding environment of the roots and the complicated growth process means that in situ and non-destructive root phenotyping face great challenges, which thus spur great research interests. The existing methods for root phenotyping are either unable to provide high-precision and high accuracy in situ detection, or they change the surrounding root environment and are destructive to root growth and health. Thus,we propose and develop an ultra-wideband microwave scanning method that uses time reversal to achieve in situ root phenotyping nondestructively. To verify the method’s feasibility, we studied an electromagnetic numerical model that simulates the transmission signal of two ultra-wideband microwave antennas. The simulated signal of roots with different shapes shows the proposed system’s capability to measure the root size in the soil. Experimental validations were conducted considering three sets of measurements with different sizes, numbers and locations, and the experimental results indicate that the developed imaging system was able to differentiate root sizes and numbers with high contrast. The reconstruction from both simulations and experimental measurements provided accurate size estimation of the carrots in the soil, which indicates the system’s potential for root imaging. Full article

Smart farming uses advanced tools and technologies such as intelligent agricultural machines, high-precision sensors, navigation systems, and sophisticated computer systems to increase the economic benefits of agriculture and reduce the associated human effort. With the increasing demands of individualized farming operations, the internet of things is a crucial technique for acquiring, monitoring, processing, and managing the agricultural resource data of precision agriculture and ecological monitoring domains. Here, an internet of things-based system scheme integrating the most recent technologies for designing a management platform for agricultural machines equipped with automatic navigation systems is proposed. Various agricultural machinery cyber-models and their corresponding sensor nodes were constructed in a pre-production phase. Three key enabling technologies—multi-optimization of agricultural machinery scheduling, development of physical architecture and software, and integration of the controller-area-network with a mobile network—were addressed to support the system scheme. A demonstrative prototype system was developed and a case study was used to validate the feasibility and effectiveness of the proposed approach. Full article

Agricultural mechanization is essential to increase farmers’ income in modern agriculture. However, the use of machinery for crop production in China is quite inefficient. To understand the obstacles limiting farmers’ use of machinery, we conducted face-to-face interview surveys with 1023 farmers (including cooperative directors, machine operators, and farmers without machines) in two major cereal-producing regions with large differences in farming scale: the North China Plain (2.7 ha per capita) and the Northeast China Plain (12.8 ha per capita). The results revealed that farmers in both regions had strong will to use machines. The obstacle preventing farmers from buying machines was the lack of machinery training in the Northeast China Plain and land fragmentation in the North China Plain. Among different farmer groups, land fragmentation was the main barrier for cooperative directors. Farmers without machines thought that there was lack of machinery training and that the cost of machinery purchase was high. Machine operators believed that machine maintenance was too expensive. The income and age also had an effect on the different groups of farmer. It is concluded that, to improve mechanization efficiency and stimulate farmers’ intention to use machinery, the government should make policies to encourage the merge of fragmented farmlands, provide targeted subsidies for agricultural machinery, and organize machinery training in an efficient way. Full article

The cultivation of medicinal plants represents great necessity and topicality these days, given that the pharmaceutical industry requires high quality raw materials in large quantities. Those are used for the production of food supplements/phytomedicines/medical devices or gemmo-derivatives’ products. Starting from these premises, this present study aimed to culture common batches of different associations of medicinal plants in order to quantify the fabrication of plant products, but also to observe possible changes in their internal structure, in direct correlation with the biosynthesis of active principles. The crops were monitored in 2018–2021. It was found that in all the common crops compared to the control ones, the amount of vegetable product provided was much higher (for example, the thyme-rosemary crop produced 730 g of fresh vegetable plants, compared with 540 g in the control crop; St. John’s Wort in culture with lemon balm delivered 1934 g of vegetable product, compared with 1423 g obtained from the control crop; mint was grown with lemon balm and produced a double amount of vegetable mass compared with the control crop). The presence of numerous glandular hairs in the samples from the phytosociological groups for the species from the Lamiaceae family, could explain the difference in the volatile oil content (4 mL/100 g produced by rosemary from the thyme-rosemary crop compared with 3.6 mL/100 g from the control one; 6.6 mL/100 g generated by thyme from the thyme-rosemary crop compared with 3.6 mL/100 from the control group; 2 mL/100 g of lemon balm volatile oil from the mint-lemon balm compared with 0.6 mL/100 g). The content of other types of active principles is dependent on the culture association. From results analysis it was found that in the phytosociological groups, flavones, PCAs and total polyphenols were significantly higher compared to control ones (2.4413 ± 0.1858 g flavones expressed in rutin/100 g in the thyme dried leaves from thyme-rosemary to 1.9317 ± 0.0947 g flavones produced by the control thyme; 9.9461 ± 0.8385 g PCAs expressed in chlorogenic acid/100 g for the same sample compared with 6.9709 ± 1.4921 g produced by the control batch; 11.1911 ± 0.7959 g TPC expressed in tannic acid/100 g in the thyme dried leaves from the thyme-rosemary phytosociological crop to 6.0393 ± 0.3204 g from the control one). The obtained results can be a starting point regarding the potential associations of medicinal plants in crops, in order to obtain a qualitative and quantitative vegetal mass. Full article

Today, the ‘Rojo Brillante’ persimmons undergoing prolonged storage are treated with giberellic acid, which allows the delay of the harvesting to November-December. Although during this period the fruit maintained high commercial firmness, practical experience indicates very different behavior during the posterior cold storage, depending on the harvest moment. To explain what leads to these differences, an in-depth study of the physicochemical and microstructural changes occurring in the fruit during five commercial harvest times from November to December was carried out. During this period, slight variations in firmness occurred, ranging from 48 to 40 N. Nevertheless, the fruit behavior under cold storage was strongly influenced by the harvest date, which was explained by the degradation of cell wall, cell membrane and tonoplast, mainly noted in fruit from the latest harvests. Therefore, the fruit harvested with firmness close to 48 N had a highly structured cell, which maintained firmness during cold storage for up to 90 days. The fruit harvested with 43 N presented a more degraded structure, while the fruit with initial firmness around 40 N underwent major ultrastructure cell wall and membranes modifications, which led to greater firmness loss. Therefore, the fruit firmness at harvest is decisive for its storage potential. Full article

To accurately recognize plant diseases under complex natural conditions, an improved plant disease-recognition model based on the original YOLOv5 network model was established. First, a new InvolutionBottleneck module was used to reduce the numbers of parameters and calculations, and to capture long-distance information in the space. Second, an SE module was added to improve the sensitivity of the model to channel features. Finally, the loss function ‘Generalized Intersection over Union’ was changed to ‘Efficient Intersection over Union’ to address the former’s degeneration into ‘Intersection over Union’. These proposed methods were used to improve the target recognition effect of the network model. In the experimental phase, to verify the effectiveness of the model, sample images were randomly selected from the constructed rubber tree disease database to form training and test sets. The test results showed that the mean average precision of the improved YOLOv5 network reached 70%, which is 5.4% higher than that of the original YOLOv5 network. The precision values of this model for powdery mildew and anthracnose detection were 86.5% and 86.8%, respectively. The overall detection performance of the improved YOLOv5 network was significantly better compared with those of the original YOLOv5 and the YOLOX_nano network models. The improved model accurately identified plant diseases under natural conditions, and it provides a technical reference for the prevention and control of plant diseases. Full article

The harvesting operation is a recurring task in the production of any crop, thus making it an excellent candidate for automation. In protected horticulture, one of the crops with high added value is tomatoes. However, its robotic harvesting is still far from maturity. That said, the development of an accurate fruit detection system is a crucial step towards achieving fully automated robotic harvesting. Deep Learning (DL) and detection frameworks like Single Shot MultiBox Detector (SSD) or You Only Look Once (YOLO) are more robust and accurate alternatives with better response to highly complex scenarios. The use of DL can be easily used to detect tomatoes, but when their classification is intended, the task becomes harsh, demanding a huge amount of data. Therefore, this paper proposes the use of DL models (SSD MobileNet v2 and YOLOv4) to efficiently detect the tomatoes and compare those systems with a proposed histogram-based HSV colour space model to classify each tomato and determine its ripening stage, through two image datasets acquired. Regarding detection, both models obtained promising results, with the YOLOv4 model standing out with an F1-Score of 85.81%. For classification task the YOLOv4 was again the best model with an Macro F1-Score of 74.16%. The HSV colour space model outperformed the SSD MobileNet v2 model, obtaining results similar to the YOLOv4 model, with a Balanced Accuracy of 68.10%. Full article