Actual Agricultural Production Research Articles

An Extended Method Based on the Geometric Position of Salient Image Features: Solving the Dataset Imbalance Problem in Greenhouse Tomato Growing Scenarios

Machine vision has significant advantages in a wide range of agricultural applications; however, acquiring a large number of high-quality image resources is often challenging in actual agricultural production due to environmental and equipment conditions. Therefore, crop image augmentation techniques are particularly important in crop growth analysis. In this paper, greenhouse tomato plants were used as research subjects to collect images of their different fertility stages with flowers and fruits. Due to the different durations of each fertility period, there is a significant difference in the number of images collected. For this reason, this paper proposes a method for balanced amplification of significant feature information in images based on geometric position. Through the geometric position information of the target in the image, different segmentation strategies are used to process the image and supervised and unsupervised methods are applied to perform balanced augmentation of the image, which is combined with the YOLOv7 algorithm to verify the augmentation effect. In terms of the image dataset, the mixed image dataset (Mix) is supplemented with mobile phone images on top of in situ monitoring images, with precision increased from 70.33% to 82.81% and recall increased from 69.15% to 81.25%. In terms of image augmentation, after supervised balanced amplification, the detection accuracy is improved from 70.33% to 77.29%, which is suitable for supervised balanced amplification. For the mobile phone dataset (MP), after amplification, it was found that better results could be achieved without any amplification method. The detection accuracy of the mixed dataset with different data sources matching the appropriate amplification method increased slightly from 82.81% to 83.59%, and accurate detection could be achieved when the target was shaded by the plant, and in different environments and light conditions.

Implementation of Large Language Models and Agricultural Knowledge Graphs for Efficient Plant Disease Detection

This study addresses the challenges of elaeagnus angustifolia disease detection in smart agriculture by developing a detection system that integrates advanced deep learning technologies, including Large Language Models (LLMs), Agricultural Knowledge Graphs (KGs), Graph Neural Networks (GNNs), representation learning, and neural-symbolic reasoning techniques. The system significantly enhances the accuracy and efficiency of disease detection through an innovative graph attention mechanism and optimized loss functions. Experimental results demonstrate that this system significantly outperforms traditional methods across key metrics such as precision, recall, and accuracy, with the graph attention mechanism excelling in all aspects, particularly achieving a precision of 0.94, a recall of 0.92, and an accuracy of 0.93. Furthermore, comparative experiments with various loss functions further validate the effectiveness of the graph attention loss mechanism in enhancing model performance. This research not only advances the application of deep learning in agricultural disease detection theoretically but also provides robust technological tools for disease management and decision support in actual agricultural production, showcasing broad application prospects and profound practical value.

Identifying rice field weeds from unmanned aerial vehicle remote sensing imagery using deep learning

BackgroundRice field weed object detection can provide key information on weed species and locations for precise spraying, which is of great significance in actual agricultural production. However, facing the complex and changing real farm environments, traditional object detection methods still have difficulties in identifying small-sized, occluded and densely distributed weed instances. To address these problems, this paper proposes a multi-scale feature enhanced DETR network, named RMS-DETR. By adding multi-scale feature extraction branches on top of DETR, this model fully utilizes the information from different semantic feature layers to improve recognition capability for rice field weeds in real-world scenarios.MethodsIntroducing multi-scale feature layers on the basis of the DETR model, we conduct a differentiated design for different semantic feature layers. The high-level semantic feature layer adopts Transformer structure to extract contextual information between barnyard grass and rice plants. The low-level semantic feature layer uses CNN structure to extract local detail features of barnyard grass. Introducing multi-scale feature layers inevitably leads to increased model computation, thus lowering model inference speed. Therefore, we employ a new type of Pconv (Partial convolution) to replace traditional standard convolutions in the model.ResultsCompared to the original DETR model, our proposed RMS-DETR model achieved an average recognition accuracy improvement of 3.6% and 4.4% on our constructed rice field weeds dataset and the DOTA public dataset, respectively. The average recognition accuracies reached 0.792 and 0.851, respectively. The RMS-DETR model size is 40.8 M with inference time of 0.0081 s. Compared with three classical DETR models (Deformable DETR, Anchor DETR and DAB-DETR), the RMS-DETR model respectively improved average precision by 2.1%, 4.9% and 2.4%.DiscussionThis model is capable of accurately identifying rice field weeds in complex real-world scenarios, thus providing key technical support for precision spraying and management of variable-rate spraying systems.

Implementation and Evaluation of Spatial Attention Mechanism in Apricot Disease Detection Using Adaptive Sampling Latent Variable Network.

In this study, an advanced method for apricot tree disease detection is proposed that integrates deep learning technologies with various data augmentation strategies to significantly enhance the accuracy and efficiency of disease detection. A comprehensive framework based on the adaptive sampling latent variable network (ASLVN) and the spatial state attention mechanism was developed with the aim of enhancing the model's capability to capture characteristics of apricot tree diseases while ensuring its applicability on edge devices through model lightweighting techniques. Experimental results demonstrated significant improvements in precision, recall, accuracy, and mean average precision (mAP). Specifically, precision was 0.92, recall was 0.89, accuracy was 0.90, and mAP was 0.91, surpassing traditional models such as YOLOv5, YOLOv8, RetinaNet, EfficientDet, and DEtection TRansformer (DETR). Furthermore, through ablation studies, the critical roles of ASLVN and the spatial state attention mechanism in enhancing detection performance were validated. These experiments not only showcased the contributions of each component for improving model performance but also highlighted the method's capability to address the challenges of apricot tree disease detection in complex environments. Eight types of apricot tree diseases were detected, including Powdery Mildew and Brown Rot, representing a technological breakthrough. The findings provide robust technical support for disease management in actual agricultural production and offer broad application prospects.

Agricultural Land Suitability Assessment Based on GIS at the County Scale in Yibin, China

As agricultural land resources scarce, food security has emerged as a pressing global concern. Conducting a suitability assessment of agricultural land is crucial for agricultural resources and food security. To establish an evaluation index system, we incorporated land, water, climate, and society four dimensions of 16 indicators. And we incorporated Entropy Weight Method (EWM), an objective method, with integrated evaluation model to assess agricultural land suitability in Yibin. Based on actual agricultural production, areas of land of urban construction, rural settlement and mining and water are subtracted. The Findings found that agricultural land suitability in Yibin presented a significant spatial heterogeneity. Areas with suitability and relative suitability accounted for 57.61% of Yibin’s land areas, which suggest Yibin has an abundance of agricultural land resources. Regions moderately suitable predominantly lie in the western and southern areas, covering 32.22% of the land area. Generally suitable areas accounted for 10.03% of Yinbin’s land areas, spatial distribution of which is similar to moderately suitable regions. Moderately and generally areas are potential areas for agricultural production. And regions for unsuitable accounted for only 0.14% of land areas. Overall, results mentioned above indicate that Yibin's agricultural land resources are in good condition. Considered terrain and water sources, this study revealed that the spatial distribution of topography and annual precipitation in Yibin closely mirrored both suitable and relatively suitable regions.

Biochar alleviates the crop failure of rice production induced by low-nitrogen cultivation mode by regulating the soil microbes taxa composition.

To improve the nitrogen utilization efficiency and a series of environmental problems caused by excessive application of nitrogen fertilizer, actual agricultural production often reduced the usage ratio of nitrogen fertilizer. However, the reduction in nitrogen fertilizer not only affects the soil microenvironment but also leads to adverse effects on rice yield. Due to its unique properties, biochar can regulate soil nutrient distribution and significantly affect soil microbial community structure/functions. To further understand the effects of different levels of biochar on soil nutrient indicators, soil microorganisms and crop growth under the nitrogen-reduction condition, our experiment with four groups was set up as followed: 0%, 2.5% and 5% biochar application rates with 99kg/hm2 nitrogen fertilizer and one control group (the actual fertilizer standard used in the field:110kg/hm2) without no exogenous biochar supplement. The rice yield and soil nutrient indexes were observed, and the differences between groups were analyzed based on multiple comparisons. 16S ribosomal RNA and ITS sequencing were used to analyze the community structure of soil bacteria and fungi. Redundancy analysis was performed to obtain the correlation relationships between microbial community marker species, soil nutrient indexes, and rice yield. Path analysis was used to determine the mechanism by which soil nutrient indexes affect rice yield. The results showed that a higher application rate of biochar led to a significant increased trend in the soil pH, organic matter and total nitrogen content. In addition, a high concentration of biochar under nitrogen-reduction condition decreased the soil bacterial diversity but elevated the fungal diversity. Different concentrations of biochar resulted in these changes in the relative abundance of soil bacteria/fungi but did not alter the dominant species taxa. Taken together, appropriate usage for biochar under the nitrogen-reduction background could induce alteration in soil nutrient indicators, microbial communities and crop yields. These results provide a theoretical basis for exploring scientific, green and efficient fertilization strategies in the rice cultivation industry. Notably, the interaction relationship between rhizosphere microorganisms in rice and soil microbial taxa are not yet clear, so further research on its detailed effects on rice production is needed. In addition, the Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis for the physiological functions of the soil microbes could only predict the potential metabolic pathways. Therefore, the next-generation metagenome techonology might be performed to explore detailed metabolic differences and accurate taxa alteration at the "species" level.

Impact of Combined Exposure to Glyphosate and Diquat on Microbial Community Structure and Diversity in Lateritic Paddy Soil

Microbial communities play crucial roles in the biogeochemical cycling of many important soil elements. Pesticides are known to affect non-targeted soil microorganisms. Glyphosate (GP) and diquat (DQ), two commonly used non-selective herbicides, often co-exist in lateritic paddy soil rich in iron/aluminum oxides. However, there is limited information on their co-impact on microbial community structure and diversity in this type of soil. Here, the short-term effects of combined exposure to GP and DQ on microbial diversity and community structure shifts were investigated in lateritic paddy soil from a tropical agricultural region (Hainan, China). The combined utilization dosages of two herbicides were set in three concentrations: low concentration (1 fold of the recommended dosage), medium concentration (10 fold of the recommended dosage) and high concentration (100 fold of the recommended dosage). The structure and diversity of microbial communities were determined via 16S rRNA and ITS gene high-throughput sequencing. The results revealed that Actinobacteria and Proteobacteria were the most sensitive microbial phyla to the combined exposure of GP and DQ in lateritic paddy soil. The combined exposure to GP and DQ increased the abundance of Actinobacteria but significantly inhibited that of Proteobacteria, especially at low and medium concentrations. Compared with CK, mixed herbicide (GP + DQ) had no adverse effects on the richness of bacteria and fungi communities as well as on the diversity of bacteria communities, but it significantly decreased the diversity of fungi communities at high concentrations within 28 days. However, the effects of combined exposure to GP and DQ on soil microbial richness and diversity were not significantly different from those of separate exposure of the two herbicides. In conclusion, the combined application of GP and DQ had no more adverse effects on soil microorganisms. Therefore, these two herbicides can be used reasonably in actual agricultural production.

Does straw returning affect the root rot disease of crops in soil? A systematic review and meta-analysis

Straw returning is a sustainable way that does not destroy soil ecology in agronomic management. Some studies have found that straw returning may aggravate or reduce soilborne diseases in the past few decades. Despite the increasing number of independent studies investigated the effect of straw returning on root rot of crops, the quantitative analysis regarding the relationship between straw returning and crop root rot is still undefined. In this study, keywords co-occurrence matrix was extracted from 2489 published studies (published from 2000 to 2022, the same below) on controlling soilborne diseases of crops. The methods used for soilborne diseases prevention have shifted from chemical to biological and agricultural control since 2010. As root rot is the soilborne disease with the largest weight in keyword co-occurrence according to statistics, we further collected 531 articles focusing on crop root rot. Notably, the 531 studies are mainly distributed in the United States, Canada, China and other countries in Europe and the south and southeast of Asia, and focus on the root rot of soybean, tomato, wheat and other important grain crops or economic crops. Based on the meta-analysis of 534 measurements in 47 previous studies, we explored how 10 management factors (soil pH/texture, type/size of straw, depth/rate/cumulative amount of application, days after application, beneficial/pathogenic microorganism inoculated before application and annual N-fertilizer input) during straw returning affect root rot onset worldwide. The results showed that straw size and microorganisms inoculated before straw returning are the key factors affecting the incidence of root rot. In combination with actual agricultural production, detailed advice applicable to traditional farming system on the optimization management of straw returning was given. This study emphasized the significance of straw pretreatment and farmland management to reduce soilborne diseases during straw returning.

Статистические ежегодники стран-членов Совета экономической взаимопомощи как источник по истории сельского хозяйства Монголии

Introduction. Statistical yearbooks of COMECON (Council for Mutual Economic Assistance) member countries to have resulted from intergovernmental cooperation are still explored by researchers for various aspects of socialist Mongolia’s socioeconomic development. Goals. The article aims to characterize statistical yearbooks of COMECON member countries as a source on the history of agriculture in Mongolia. Materials and methods. The paper analyzes the mentioned statistical yearbooks published between 1970 and 1990. The main research methods employed are the historical descriptive, comparative, and analytical ones. Results. The agricultural section of yearbooks includes an impressive number of indicators grouped into various categories. The quantitative and qualitative composition of indicators remained virtually unchanged from 1970 to 1989, while the 1990 yearbook witnessed significant structural changes. Conclusions. The statistical yearbooks of COMECON member countries are consolidated statistical digests with a fairly voluminous nomenclature of indicators, extensive time coverage, and a well-defined structure. The main advantage of theirs over national statistical materials is that they grant an opportunity to conduct a comparative analysis of various indicators, including in the agricultural sector. The most important data are specific indicators and average values that can primarily attest to actual agricultural production intensification degrees in Mongolia in 1960–1989. All this characterizes statistical yearbooks of COMECON member countries as a very important and useful source on the history of agriculture in Mongolia.

Farmland Transfer, Scale Management and Economies of Scale Assessment: Evidence from the Main Grain-Producing Shandong Province in China

Promoting farmland transfer through the farmland rental market is an essential instrument to achieve the centralized scale operation of farmland in China. However, few studies have explored or verified the economies of scale after land concentration. This study uses face-to-face interviews to randomly select 395 farmers engaged in land scale management in 68 villages of 11 cities in Shandong Province. Based on these data, a production cost function model is used to measure the cost elasticity of farmland scale management and further empirically test whether farmland scale management can achieve economies of scale in actual agricultural production. The empirical results show that the development of farmland scale operation can reduce production costs and realize the economy of scale. In other words, the scale of farmland management still has the possibility of further expansion. More importantly, we find the operation cost of farmland is significantly influenced by the price of farmland transfer, fixed capital input and labor input, especially the scale operator with rich planting experience and good cultural quality is an important human capital stock, which can make significant contribution to reducing production cost and developing the positive effect of scale operation. To promote farmland scale management in China, we should pay attention to reducing the transferring cost and transaction cost of farmland by building the land transfer market trading platform, increasing the subsidies for farmland transfer and developing agricultural mechanization, which is helpful to improve agricultural productivity and realize the scale economy. This research can provide a reference for rational land scale management and land use policymaking.

Rape straw application facilitates Se and Cd mobilization in Cd-contaminated seleniferous soils by enhancing microbial iron reduction

Many naturally seleniferous soils are faced with Cd contamination problem, which severely limits crop cultivation in these areas. Straw returning has been widely applied in agricultural production due to its various benefits to soil physicochemical properties, soil fertility, and crops yield. However, effects of straw application on the fates of Se and Cd in Cd-contaminated seleniferous soils remain largely unclear. Therefore, the effects of straw application on the fates of Se and Cd in Cd-contaminated seleniferous soils were investigated in this study. The results showed that iron reduction driven by Clostridium and Anaeromyxbacter was responsible for the variations in Se and Cd fates in soil. Straw application respectively increased the gene copy numbers of Clostridium and Anaeromyxbacter by 19.5–56.3% and 33.6–39.8%, thus promoting iron reductive dissolution, eventually resulting in a high release amount of Se and Cd from Fe(III) (oxyhydr) oxides. Under reducing conditions, the released Cd was adsorbed by the newly formed metal sulfides or reacted with sulfides to generate CdS precipitates. Straw application decreased the soil exchangeable Se and soil exchangeable Cd concentration during flooding phase. However, straw application significantly increased Se/Cd in soil solution which had the highest bioavailability during flooding. In addition, straw application increased soil exchangeable Se concentration, but it had no significant effects on soil exchangeable Cd concentration after soil drainage. Taken together, straw application increased Se bioavailability and Cd mobility. Therefore, straw application is an effective method for improving Se bioavailability, but it is not suitable for the application to Cd-contaminated paddy soils. In the actual agricultural production, straw could be applied in seleniferous soils to improve Se bioavailability. At the same time, straw application should be cautious to avoid the release of Cd from Cd-contaminated soil.

Evaluation of Soil and Irrigation Water Quality in Caohai Lakeside Zone

Due to the rapid population growth and over-application of fertilizers in the Caohai surrounding farmlands, controlling the non-point source pollution in the Caohai Lakeside Zone is significant for the local ecology balance and human health safety. A total of 54 soil and 24 irrigation water samples were collected in the Caohai Lakeside Zone to evaluate the current status of soil fertility and heavy metal pollution by measuring soil and water heavy metal content and soil physicochemical properties such as soil total nitrogen, total phosphorus, organic matter, and soil pH. These results showed that the total amounts of organic matter, nitrogen, phosphorus, potassium, and effective nutrient content in the Caohai Lakeside Zone were all at a rich level according to Chinese soil nutrient grade standard; the content of lead (Pb), chromium (Cr), cadmium (Cd), mercury (Hg), and arsenic (As) in the soil all exceeds the safety standard of Chinese classification of soil environmental quality assessment. In addition, the over-standard rates of Cd and Cr were 24.1% and 14.8%, respectively. On the whole, the comprehensive quality index (CQI) of the soil in the Caohai Lakeside Zone is 6.48, which is attributed to the heavy Cd pollution and the good soil fertility. The heavy metal content of the irrigation water met the requirements of irrigation water quality. It is feasible to use the irrigation water of Caohai Lake in the actual agricultural production process. Therefore, in terms of the comprehensive management and pollution control of Caohai, special attention should be paid to the control of Cd, Pb, As, and Hg pollution sources.

Fused-Deep-Features Based Grape Leaf Disease Diagnosis

Rapid and accurate grape leaf disease diagnosis is of great significance to its yield and quality of grape. In this paper, aiming at the identification of grape leaf diseases, a fast and accurate detection method based on fused deep features, extracted from a convolutional neural network (CNN), plus a support vector machine (SVM) is proposed. In the research, based on an open dataset, three types of state-of-the-art CNN networks, three kinds of deep feature fusion methods, seven species of deep feature layers, and a multi-class SVM classifier were studied. Firstly, images were resized to meet the input requirements of the CNN network; then, the deep features of the input images were extracted via the specific deep feature layer of the CNN network. Two kinds of deep features from different networks were then fused using different fusion methods to increase the effective classification feature information. Finally, a multi-class SVM classifier was trained with the fused deep features. The experimental results on the open dataset show that the fused deep features with any kind of fusion method can obtain a better classification performance than using a single type of deep feature. The direct concatenation of the Fc1000 deep feature extracted from ResNet50 and ResNet101 can achieve the best classification result compared with the other two fusion methods, and its F1 score is 99.81%. Furthermore, the SVM classifier trained using the proposed method can achieve a classification performance comparable to that of using the CNN model directly, but the training time is less than 1 s, which has an advantage over spending tens of minutes training a CNN model. The experimental results indicate that the method proposed in this paper can achieve fast and accurate identification of grape leaf diseases and meet the needs of actual agricultural production.

Effect of Vegetation Restoration on Soil Hydrology in Karst Area of Southwest China: Inspiration from Barrel Planting Experiments

The purpose of this study was to explore the effects of different vegetation restoration types on soil hydrology characteristics in the Karst Plateau Gorge and to clarify the soil moisture (θ) use characteristics. A barrel experiment was conducted to monitor θ and the water potential (Ψ) of three vegetation types (Zanthoxylum bungeanum (ZB), Zea mays L. (ZM), and Sophora tonkinensis (ST), Abandoned land (AL)) was used as a control to explore θ use conditions of each vegetation type. A larger surface permeability led to lower moisture storage. The soil moisture storage showed the law of ZM > ST > AL > ZB. The soil moisture storage also had obvious characteristics in dry-wet seasons. As a typical drought-tolerant crop, ZB responded more obviously to rainfall and had the highest effective replenishment amount and efficiency. Two processes were clearly involved in decreasing soil moisture, which could be divided into three stages of changes: a consumption period (CP), a moisture supplying period (SP), and a relatively stable period (RSP). CP occurred primarily in November to April, when θ was prone to water stress and required proper artificial replenishment. SP was characterized by limited rainfall replenishment in January and May, which significantly increased θ. During the rainy season, corresponding with RSP, θ fluctuated within a relatively stable range. At the end of the CP, the water shortage was more severe. In actual agricultural production, attention should be given to reasonable artificial recharge. This research aims to provide a theoretical basis for karst θ management.

Optimization of autonomous driving state control of low energy consumption pure electric agricultural vehicles based on environmental friendliness.

The distributed driven agricultural vehicle platform studied in this paper is a basic carrying platform that can be used for a variety of agricultural activities (pesticide spraying, weeding, and other plant protection operations). It does not need to rely on traditional fuel engine to drive but is driven by electric power, which is more environmentally friendly and conducive to the development of ecological agriculture in the future. In this paper, according to the driving and structure characteristics of distributed driving electric agricultural vehicles, in order to ensure the vehicle trajectory tracking ability, under conditions of agricultural operation, the driving attitude and drive coordination control of electric agricultural vehicles are taken as the key research objectives. In this way, the driving path of the vehicle platform can be accurately controlled during the operation of pesticide spraying and herbicide, so as to reduce the excessive use or misuse of pesticides and herbicides and greatly reduce the field pollution. Considering the specific driving environment as well as complex and changeable motion patterns of agricultural vehicles, when a single motion model is used to track and estimate the driving state, there will be low filtering accuracy or even loss of the target during vehicle maneuvering. In this paper, interactive multiple model (IMM) algorithm is combined with extended Kalman filter to effectively track changes of the target's motion mode, thereby avoiding low filtering accuracy or serious state estimation inaccuracy. Finally, through the distributed electric drive agricultural operation experimental platform developed by the research group, the working conditions close to the actual agricultural production activities was set up according to the needs of actual agricultural production activities in this paper, and applicability and accuracy of the algorithm state estimation are verified by experiments.

A novel spectroscopy-based method using monopole antenna for measuring soil water content

Soil water content measurement is indispensable to precision agriculture. A novel soil water content measuring method based on frequency scanning is proposed in this research. The measuring system consists of a monopole antenna to emit and receive electromagnetic waves as well as a network analyzer. The soil equivalent circuit model is presented and its spectroscopy is obtained by comparing incident and reflected waves. The results indicate that the resonance frequency shifts towards the low frequency region as soil moisture increases, therefore soil water content can be obtained simply by detecting resonance frequency. The linear fitting curve has a determination coefficient of 0.9834 and a root mean square error of 1.4%. The measuring range from 5.0% to 37.4% is wide enough for actual agricultural production. The method offers a productive solution for soil water content measurement and provides an idea to sense other soil characteristics.

Predicting copper contamination in wheat canopy during the full growth period using hyperspectral data.

The rapid and efficient determination of heavy metal content in food crops is essential for human health and environmental protection. The use of hyperspectral data has become a popular way to predict heavy metal content in plants; however, many challenges remain. One challenge is that lab conditions differ from actual agricultural production conditions. Another challenge is that spectral data characteristics are not universally applicable to all situations. Therefore, in this study, the field test method was adopted to conduct experiments during the full growth period of wheat, and the spectrum data of wheat canopy were processed by the first derivative method to screen-sensitive spectral bands as the basis for the prediction model of the copper content in wheat. The results showed that the copper content increased with an increase in the soil copper content, and there were dissimilar subtle differences in the spectral reflectance of wheat canopy under different stressed soil copper concentrations; sensitive spectral indices and wavelengths were screened based on good correlation with the copper content in the wheat canopy. Different optimal predicting models in different periods were built and verified. The established linear regression models, which were based on NDVI/SIPI and W728, were the most suitable predicting models during the tillering stage with R2 = 0.669 and 0.818; Rg, W741, and multiple bands were the most suitable predicting models during the jointing stage with R2 = 0.548, 0.830, and 0.868; the optimal model during the heading stage was based on W480 (R2 = 0.625). This study demonstrated that the constructed models had good potential for estimating the copper content in wheat leaves during full growth periods, and this method had the potential to be applied to the actual agricultural production process.

The Role of Collective Food Identity in Local Food Demand

AbstractDespite a proliferation of articles focused on consumer willingness-to-pay for locally produced foods, few studies have systemically considered the role of food identity in the price premium of local foods. This article uses primary and secondary data to fill that gap. Using data from 9,329 U.S. households, we identify what foods each state considers a part of their collective food identity. We then compare each state's collective food identity to the actual production within the state. Finally, we utilize a discrete choice experiment (N = 484) focused on a state with one such collective food identity to test whether a consumer's preference for local foods is a function of the collective food identity within the state. Results from the open-ended survey suggest that each state connects with their food production in unique ways and that not all state identities are consistent with actual agricultural production. Results from the discrete choice experiment suggest that consumers are more willing to pay a premium for the local food that is a part of their local food identity than for a similar product that is not a part of their local food identity. Our findings imply that future research on consumer preferences for local food systems should consider the relationship between food culture and geography.

Differing trade elasticities for intra‐ and international distances: A gravity approach

AbstractUsing the gravity model of trade, I estimate the impact that within‐country transportation distance has on international trade levels. Combining multiple data sets, I create a measure for the distance U.S. agricultural goods travel before leaving the country. In order to account for endogeneity in production location choice, I use potential agricultural production as an instrument for actual agricultural production. I find that internal distance is statistically significant and large in magnitude. A 10% reduction in the distance a good travels within the exporting country increases trade by roughly 3%.

MDFC–ResNet: An Agricultural IoT System to Accurately Recognize Crop Diseases

Crop disease diagnosis is an essential step in crop disease treatment and is a hot issue in agricultural research. However, in agricultural production, identifying only coarse-grained diseases of crops is insufficient because treatment methods are different in different grades of even the same disease. Inappropriate treatments are not only ineffective in treating diseases but also affect crop yield and food safety. We combine IoT technology with deep learning to build an IoT system for crop fine-grained disease identification. This system can automatically detect crop diseases and send diagnostic results to farmers. We propose a multidimensional feature compensation residual neural network (MDFC-ResNet) model for fine-grained disease identification in the system. MDFC-ResNet identifies from three dimensions, namely, species, coarse-grained disease, and fine-grained disease and sets up a compensation layer that uses a compensation algorithm to fuse multidimensional recognition results. Experiments show that the MDFC-ResNet neural network has better recognition effect and is more instructive in actual agricultural production activities than other popular deep learning models.