A time-varying, nonlinear soil-plant system contains many unknown elements that can be quantified based on analytical methodologies. Artificial Neural Networks (ANNs) are a widely used mathematical computing, modelling, and predicting method that estimates unknown values of variables from known values of others. This paper aims to simulate relationship between soil moisture, bulk density, porosity ratio, depth, and penetration resistance and to estimate soil penetration resistance with the help of ANNs. For this aim, the Generalized Regression Neural network (GRNN) and Radial Basis Function (RBF) models were developed and compared for the estimation of soil penetration resistance values in MATLAB. A dataset of 153 samples was collected from experimental field. From the 153 data, 102 data (33%) were selected for training and the remaining 51 data (67%) were used for testing. The estimation process was implemented 10 replications using randomly selected testing and training data. Mean Squared Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) were used to evaluate estimation accuracy on the developed ANN methods. Based on MSE, RMSE, MAE and Standard Deviation (SD), statistical results showed that the GRNN modelling presented better results than the RBF model in predicting soil penetration resistance success.
{"title":"Comparison of two different artificial neural network models for prediction of soil penetration resistance","authors":"I. Ünal, Ö. Kabaş, S. Sözer","doi":"10.4081/jae.2023.1550","DOIUrl":"https://doi.org/10.4081/jae.2023.1550","url":null,"abstract":"A time-varying, nonlinear soil-plant system contains many unknown elements that can be quantified based on analytical methodologies. Artificial Neural Networks (ANNs) are a widely used mathematical computing, modelling, and predicting method that estimates unknown values of variables from known values of others. This paper aims to simulate relationship between soil moisture, bulk density, porosity ratio, depth, and penetration resistance and to estimate soil penetration resistance with the help of ANNs. For this aim, the Generalized Regression Neural network (GRNN) and Radial Basis Function (RBF) models were developed and compared for the estimation of soil penetration resistance values in MATLAB. A dataset of 153 samples was collected from experimental field. From the 153 data, 102 data (33%) were selected for training and the remaining 51 data (67%) were used for testing. The estimation process was implemented 10 replications using randomly selected testing and training data. Mean Squared Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) were used to evaluate estimation accuracy on the developed ANN methods. Based on MSE, RMSE, MAE and Standard Deviation (SD), statistical results showed that the GRNN modelling presented better results than the RBF model in predicting soil penetration resistance success.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":" 31","pages":""},"PeriodicalIF":1.8,"publicationDate":"2023-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139144367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To ensure that a variable-rate spray (VRS) system can perform unmanned aerial spray in accordance with a prescription map at different flight speeds, we examine in this paper such significant factors as the response time of the VRS system and the pressure fluctuation of the nozzle during the variable-rate spraying process. The VRS system uses a lag compensation algorithm (LCA) to counteract the droplet deposition position lag caused by the system response delay. In addition, pulse width modulated (PWM) solenoid valves are used for controlling the flowrates of the nozzles on the variable-rate spray system, and a mathematical model was constructed for the spray rate (L min-1) and the relative proportion of time (duty cycle) each solenoid valve is open. The pressure drop and solenoid valve response time at different duty cycles (50%~90%) were measured by indoor experiments. Meanwhile, the lag distance (LD), spray accuracy, and droplet deposition characteristics of the VRS system were tested by conducting outdoor experiments at different flight speeds (4m s-1, 5m s-1, 6m s-1). The results show that LCA can effectively reduce the lag distance. The lag distance (LD) values of the VRS system with LCA ranged from -0.27 to 0.78m with an average value of 0.32m, while without LCA, the LD values increased to 3.5~4.3m with an average value of 3.87m. The overall spray position accuracy was in the range of 91.56%~97.32%. Furthermore, the spray coverage and deposition density, determined using water sensitive paper (WSP), were used to evaluate the spray application performance taking into account the spray volume applied. The VRS system can provide the most suitable spray volumes for insecticide and fungicide plant protection products. Based on a prescription map, the optimized VRS system can achieve accurate pesticide spraying as well as desirable spray coverage and deposition density.
{"title":"Variable-rate spray system for unmanned aerial applications using lag compensation algorithm and pulse width modulation spray technology","authors":"Zhongkuan Wang, Sheng Wen, Yubin Lan, Yue Liu, Yingying Dong","doi":"10.4081/jae.2023.1547","DOIUrl":"https://doi.org/10.4081/jae.2023.1547","url":null,"abstract":"To ensure that a variable-rate spray (VRS) system can perform unmanned aerial spray in accordance with a prescription map at different flight speeds, we examine in this paper such significant factors as the response time of the VRS system and the pressure fluctuation of the nozzle during the variable-rate spraying process. The VRS system uses a lag compensation algorithm (LCA) to counteract the droplet deposition position lag caused by the system response delay. In addition, pulse width modulated (PWM) solenoid valves are used for controlling the flowrates of the nozzles on the variable-rate spray system, and a mathematical model was constructed for the spray rate (L min-1) and the relative proportion of time (duty cycle) each solenoid valve is open. The pressure drop and solenoid valve response time at different duty cycles (50%~90%) were measured by indoor experiments. Meanwhile, the lag distance (LD), spray accuracy, and droplet deposition characteristics of the VRS system were tested by conducting outdoor experiments at different flight speeds (4m s-1, 5m s-1, 6m s-1). The results show that LCA can effectively reduce the lag distance. The lag distance (LD) values of the VRS system with LCA ranged from -0.27 to 0.78m with an average value of 0.32m, while without LCA, the LD values increased to 3.5~4.3m with an average value of 3.87m. The overall spray position accuracy was in the range of 91.56%~97.32%. Furthermore, the spray coverage and deposition density, determined using water sensitive paper (WSP), were used to evaluate the spray application performance taking into account the spray volume applied. The VRS system can provide the most suitable spray volumes for insecticide and fungicide plant protection products. Based on a prescription map, the optimized VRS system can achieve accurate pesticide spraying as well as desirable spray coverage and deposition density.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"56 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135869515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anthropogenic activities have adverse impacts on productive lands around coastal zones due to rapid developments. Assessment of land use and land cover (LULC) changes provides better understanding of the process for conservation of such vulnerable ecosystems. Alanya is one of the most popular tourism hotspots in Mediterranean coast of Turkey, and even though the city faced with severe LULC changes after mid-80s due to tourism-related investments, limited number of studies has conducted in the area The study aimed to determine short-term and long-term LULC changes and effects of residential development process on agricultural lands using six Landsat imageries acquired between 1984 and 2017, and presented the first attempt of future simulation in the area. Average annual conversions (AAC) (ha) calculated to assess magnitudes of annual changes in six different periods. AACs used to calculate area demands for LULC2030 and LULC2050, whereby annual conversions from different periods were multiplied by number of years between 2017, 2030 and 2050 for each scenario. Finally, optimistic and pessimistic scenarios for agricultural lands are simulated using FLUS model. Accordingly, agricultural lands decreased from 53.9% to 31.4% by 22.5% in 33 years, and predicted to change between 19.50% and 24.63% for 2030, 1.07% and 14.10% for 2050, based on pessimistic and optimistic scenarios, respectively.
{"title":"Monitoring and multi-scenario simulation of agricultural land changes using Landsat imageries and FLUS model on coastal Alanya","authors":"Melis Inalpulat","doi":"10.4081/jae.2023.1548","DOIUrl":"https://doi.org/10.4081/jae.2023.1548","url":null,"abstract":"Anthropogenic activities have adverse impacts on productive lands around coastal zones due to rapid developments. Assessment of land use and land cover (LULC) changes provides better understanding of the process for conservation of such vulnerable ecosystems. Alanya is one of the most popular tourism hotspots in Mediterranean coast of Turkey, and even though the city faced with severe LULC changes after mid-80s due to tourism-related investments, limited number of studies has conducted in the area The study aimed to determine short-term and long-term LULC changes and effects of residential development process on agricultural lands using six Landsat imageries acquired between 1984 and 2017, and presented the first attempt of future simulation in the area. Average annual conversions (AAC) (ha) calculated to assess magnitudes of annual changes in six different periods. AACs used to calculate area demands for LULC2030 and LULC2050, whereby annual conversions from different periods were multiplied by number of years between 2017, 2030 and 2050 for each scenario. Finally, optimistic and pessimistic scenarios for agricultural lands are simulated using FLUS model. Accordingly, agricultural lands decreased from 53.9% to 31.4% by 22.5% in 33 years, and predicted to change between 19.50% and 24.63% for 2030, 1.07% and 14.10% for 2050, based on pessimistic and optimistic scenarios, respectively.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"58 9","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135863310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In order to improve the efficiency of robots picking apples in challenging orchard environments, a method for precisely detecting apples and planning the picking sequence is proposed. Firstly, the EfficientFormer network serves as the foundation for YOLOV5, which uses the EF-YOLOV5s network to locate apples in difficult situations. Meanwhile, the Soft Non-Maximum Suppression (NMS) algorithm is adopted to achieve accurate identification of overlapping apples. Secondly, the adjacently identified apples are automatically divided into different picking clusters by the improved density-based spatial clustering of applications with noise (DBSCAN). Finally, the order of apple harvest is determined to guide the robot to complete the rapid picking, according to the weight of the Gauss distance weight combined with the significance level. In the experiment, the average precision of this method is 98.84%, which is 4.3% higher than that of YOLOV5s. Meanwhile, the average picking success rate and picking time are 94.8% and 2.86 seconds, respectively. Compared with sequential and random planning, the picking success rate of the proposed method is increased by 6.8% and 13.1%, respectively. The research proves that this method can accurately detect apples in complex environments and improve picking efficiency, which can provide technical support for harvesting robots.
{"title":"Apple recognition and picking sequence planning for harvesting robot in the complex environment","authors":"Wei Ji, Tong Zhang, Bo Xu, Guozhi He","doi":"10.4081/jae.2023.1549","DOIUrl":"https://doi.org/10.4081/jae.2023.1549","url":null,"abstract":"In order to improve the efficiency of robots picking apples in challenging orchard environments, a method for precisely detecting apples and planning the picking sequence is proposed. Firstly, the EfficientFormer network serves as the foundation for YOLOV5, which uses the EF-YOLOV5s network to locate apples in difficult situations. Meanwhile, the Soft Non-Maximum Suppression (NMS) algorithm is adopted to achieve accurate identification of overlapping apples. Secondly, the adjacently identified apples are automatically divided into different picking clusters by the improved density-based spatial clustering of applications with noise (DBSCAN). Finally, the order of apple harvest is determined to guide the robot to complete the rapid picking, according to the weight of the Gauss distance weight combined with the significance level. In the experiment, the average precision of this method is 98.84%, which is 4.3% higher than that of YOLOV5s. Meanwhile, the average picking success rate and picking time are 94.8% and 2.86 seconds, respectively. Compared with sequential and random planning, the picking success rate of the proposed method is increased by 6.8% and 13.1%, respectively. The research proves that this method can accurately detect apples in complex environments and improve picking efficiency, which can provide technical support for harvesting robots.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"56 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135813816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Traditional techniques for estimating the weight of clusters in a winery, generally consist of manually counting the variety of clusters per vine, and scaling by means of the entire variety of vines. This method can be arduous, costly, and its accuracy is dependent on the scale of the sample. To overcome these problems, hybrid approaches of Computer Vision (CV), Deep Learning (DL) and Machine Learning (ML) based vineyard yield prediction systems are proposed. Self-prepared datasets are used for comparative analysis of 2D and 3D yield prediction systems for vineyards. DL-based approach for segmentation operation on an RGB-D image dataset created with the D435I camera is used along with the ML-based weight prediction technique of grape clusters present in the single image is employed using these datasets. A comparative analysis of the DL-based Keras regression model and various ML-based regression models for the weight prediction task is taken into account, and finally a prediction model is proposed to estimate the yield of the entire vineyard. The analysis shows improved performance with the 3D vineyard yield prediction system compared to the 2D vineyard yield prediction system with grape cluster segmentation pixel accuracy upto 94.81% and yield prediction accuracy upto 99.58%.
{"title":"Comparative analysis of 2D and 3D vineyard yield prediction system using artificial intelligence","authors":"Dhanashree Barbole, Parul M. Jadhav","doi":"10.4081/jae.2023.1545","DOIUrl":"https://doi.org/10.4081/jae.2023.1545","url":null,"abstract":"Traditional techniques for estimating the weight of clusters in a winery, generally consist of manually counting the variety of clusters per vine, and scaling by means of the entire variety of vines. This method can be arduous, costly, and its accuracy is dependent on the scale of the sample. To overcome these problems, hybrid approaches of Computer Vision (CV), Deep Learning (DL) and Machine Learning (ML) based vineyard yield prediction systems are proposed. Self-prepared datasets are used for comparative analysis of 2D and 3D yield prediction systems for vineyards. DL-based approach for segmentation operation on an RGB-D image dataset created with the D435I camera is used along with the ML-based weight prediction technique of grape clusters present in the single image is employed using these datasets. A comparative analysis of the DL-based Keras regression model and various ML-based regression models for the weight prediction task is taken into account, and finally a prediction model is proposed to estimate the yield of the entire vineyard. The analysis shows improved performance with the 3D vineyard yield prediction system compared to the 2D vineyard yield prediction system with grape cluster segmentation pixel accuracy upto 94.81% and yield prediction accuracy upto 99.58%.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"547 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136068150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lei Liu, Xianliang Wang, Xiaokang Zhong, Xiangcai Zhang, Yuanle Geng, Hua Zhou, Tao Chen
The strip rotary tillage method effectively reduces the occurrence of straw clogging and creates a favorable seed bed environment. However, the mixture of crushed straw and soil in the seeding area results in inadequate seed-soil contact following compaction by the press wheels. A chisel-type opener furrow side pick-up blade was proposed to improve seed-soil contact by picking up wet soil from the furrow's side. The discrete element method was used to investigate the impact of earth blade surface parameters on soil dynamics. The key factors of the blade, including forward velocity, endpoint tangent angle, and angle of soil entry, were determined through theoretical analysis. Soil cover thickness and straw ratio in the seed furrow were evaluated using orthogonal rotation regression tests. The results show that the endpoint tangent angle and angle of soil entry have the greatest influence on soil cover thickness, while the angle of soil entry has the greatest influence on the straw ratio. The optimal values for the forward velocity, endpoint tangent angle, and angle of soil entry are 4.86 km/h, 107.17°, and 5.46°, respectively, resulting in a soil cover thickness of 40 mm and a straw ratio of 21.46%. Confirmatory soil bin tests showed similar results, with a soil cover thickness of 40.4 mm and a straw ratio of 18.03%. These results provide a viable solution for improving seed-soil contact after strip rotary tillage planter seeding.
{"title":"Design and experiment of furrow side pick-up soil blade for wheat strip-till planter using the discrete element method","authors":"Lei Liu, Xianliang Wang, Xiaokang Zhong, Xiangcai Zhang, Yuanle Geng, Hua Zhou, Tao Chen","doi":"10.4081/jae.2023.1546","DOIUrl":"https://doi.org/10.4081/jae.2023.1546","url":null,"abstract":"The strip rotary tillage method effectively reduces the occurrence of straw clogging and creates a favorable seed bed environment. However, the mixture of crushed straw and soil in the seeding area results in inadequate seed-soil contact following compaction by the press wheels. A chisel-type opener furrow side pick-up blade was proposed to improve seed-soil contact by picking up wet soil from the furrow's side. The discrete element method was used to investigate the impact of earth blade surface parameters on soil dynamics. The key factors of the blade, including forward velocity, endpoint tangent angle, and angle of soil entry, were determined through theoretical analysis. Soil cover thickness and straw ratio in the seed furrow were evaluated using orthogonal rotation regression tests. The results show that the endpoint tangent angle and angle of soil entry have the greatest influence on soil cover thickness, while the angle of soil entry has the greatest influence on the straw ratio. The optimal values for the forward velocity, endpoint tangent angle, and angle of soil entry are 4.86 km/h, 107.17°, and 5.46°, respectively, resulting in a soil cover thickness of 40 mm and a straw ratio of 21.46%. Confirmatory soil bin tests showed similar results, with a soil cover thickness of 40.4 mm and a straw ratio of 18.03%. These results provide a viable solution for improving seed-soil contact after strip rotary tillage planter seeding.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"160 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136071044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chengcheng Ma, Shujuan Yi, Guixiang Tao, Yifei Li, Hanwu Liu
Under the condition of high-speed sowing (12-16 km/h), due to the high rotational speeds of the seed disk, the seeds leave the disk at an excessively high speed, which challenges the seed-picking capacity of the belt-type high-speed seed guide device. In this paper, the theory of seed-picking mech-anism is analyzed, and performance optimization tests are completed to further improve the operation effect of the seeder. The mechanical model of seed picking was established through the force analysis of seeds. The influence of vacuum degree, feeder wheel rotation speed, and seed picking angle on seed picking quality and the parameter range of each factor were obtained by single factor test. A three-factor five-level quadratic orthogonal rotation combination test was performed, and the test re-sults were refined and evaluated. The test factors used were vacuum degree, feeder wheel rotation speed, and seed picking angle. The test indexes used were the seed picking rate, re- picking rate, and miss- picking rate. According to the results, the seed picking rate was 99.89%, the re-picking rate was 0, and the miss-picking rate was 0.11% when the vacuum degree was 6.89KPa, the feeder wheel rota-tion speed was 568.95rpm, and the seed picking angle was 7.6°.
{"title":"Theoretical analysis and experiment of seed-picking mechanism of belt high-speed seed-guiding device for corn","authors":"Chengcheng Ma, Shujuan Yi, Guixiang Tao, Yifei Li, Hanwu Liu","doi":"10.4081/jae.2023.1543","DOIUrl":"https://doi.org/10.4081/jae.2023.1543","url":null,"abstract":"Under the condition of high-speed sowing (12-16 km/h), due to the high rotational speeds of the seed disk, the seeds leave the disk at an excessively high speed, which challenges the seed-picking capacity of the belt-type high-speed seed guide device. In this paper, the theory of seed-picking mech-anism is analyzed, and performance optimization tests are completed to further improve the operation effect of the seeder. The mechanical model of seed picking was established through the force analysis of seeds. The influence of vacuum degree, feeder wheel rotation speed, and seed picking angle on seed picking quality and the parameter range of each factor were obtained by single factor test. A three-factor five-level quadratic orthogonal rotation combination test was performed, and the test re-sults were refined and evaluated. The test factors used were vacuum degree, feeder wheel rotation speed, and seed picking angle. The test indexes used were the seed picking rate, re- picking rate, and miss- picking rate. According to the results, the seed picking rate was 99.89%, the re-picking rate was 0, and the miss-picking rate was 0.11% when the vacuum degree was 6.89KPa, the feeder wheel rota-tion speed was 568.95rpm, and the seed picking angle was 7.6°.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"9 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136103489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Deep convolutional neural network (DCNN) has recently made significant strides in classification and recognition of rice leaf disease. The majority of classification models perform disease image recognitions using a collocation patterns including pooling layers, convolutional layers, and fully connected layers, followed by repeating this structure to complete depth increase. However, the key information of the lesion area is locally limited. That is to say, in the case of only performing feature extraction according to the above-mentioned model, redundant and low-correlation image feature information with the lesion area will be received, resulting in low accuracy of the model. For improvement of the network structure and accuracy promotion, here we proposed a double-branch DCNN (DBDCNN) model with a convolutional block attention module (CBAM). The results show that the accuracy of the classic models VGG-16, ResNet-50, ResNet50+CBAM, MobileNet-V2, GoogLeNet, EfficientNet-B1 and Inception-V2 is lower than the accuracy of the model in this paper (98.73%). Collectively, the DBDCNN model here we proposed might be a better choice for classification and identification of rice leaf diseases in the future, based on its novel identification strategy for crop disease diagnosis.
{"title":"Double-branch deep convolutional neural network-based rice leaf diseases recognition and classification","authors":"Xiong Bi, Hongchun Wang","doi":"10.4081/jae.2023.1544","DOIUrl":"https://doi.org/10.4081/jae.2023.1544","url":null,"abstract":"Deep convolutional neural network (DCNN) has recently made significant strides in classification and recognition of rice leaf disease. The majority of classification models perform disease image recognitions using a collocation patterns including pooling layers, convolutional layers, and fully connected layers, followed by repeating this structure to complete depth increase. However, the key information of the lesion area is locally limited. That is to say, in the case of only performing feature extraction according to the above-mentioned model, redundant and low-correlation image feature information with the lesion area will be received, resulting in low accuracy of the model. For improvement of the network structure and accuracy promotion, here we proposed a double-branch DCNN (DBDCNN) model with a convolutional block attention module (CBAM). The results show that the accuracy of the classic models VGG-16, ResNet-50, ResNet50+CBAM, MobileNet-V2, GoogLeNet, EfficientNet-B1 and Inception-V2 is lower than the accuracy of the model in this paper (98.73%). Collectively, the DBDCNN model here we proposed might be a better choice for classification and identification of rice leaf diseases in the future, based on its novel identification strategy for crop disease diagnosis.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"161 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136104265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ugo Lazzaro, Caterina Mazzitelli, Benedetto Sica, Paola Di Fiore, Nunzio Romano, Paolo Nasta
Two pedotransfer functions (PTFs) are available in the literature enabling the soil water retention function (WRF) to be estimated from knowledge of the soil particle-size distribution (PSD), oven-dry soil bulk density (b), and saturated soil water content (s): i) the Arya and Heitman model (PTF-AH) and ii) the Mohammadi and Vanclooster model (PTF-MV). These physico-empirical PTFs rely on the hypothesis of shape similarity between PSD and WRF, and do not require the calibration of the input parameters. In the first stage, twenty-seven PSD models were evaluated using 4,128 soil samples collected in Campania (southern Italy). These models were ranked according to the root mean square residuals (RMSR), corrected Akaike Information Criterion (AICc), and adjusted coefficient of determination (R2adj). In the second stage, three subsets of PSD and WRF data (DS-1, DS-2, and DS-3), comprising 282 soil samples, were used to evaluate the two PTFs using the best three PSD models selected in the first stage. The hypothesis of shape similarity was assumed as acceptable only when the RMSR value was lower than the field standard deviation of the WRFs (*), which is viewed as a tolerance threshold and computed from the physically-based scaling approach proposed by Kosugi and Hopmans (1998). In the first study area (DS-1), characterized by a fairly uniform, loamy textured volcanic soil, the PTF-AH outperformed the PTF-MV and both PTFs provided reasonable performance within the acceptance threshold (i.e., RMSR < *). In the other two heterogeneous field sites (DS-2 and DS-3, characterized by soil textural classes that span from clay and clay-loam to loam and even sandy-loam soils), the PTF-MV (with 3% to 6% RMSR surpassing *) outperformed the PTF-AH (with 8% to 30% RMSR surpassing *) and the majority of RMSR values were larger than those obtained in the original studies. The mean relative error (MRE) revealed that the PTF-MV systematically underestimates the measured WRFs, whereas the PTF-AH provided negative MRE values indicating an overall overestimation. The outcomes of our study provide a critical evaluation when using calibration-free PTFs to predict WRFs over large areas.
{"title":"On evaluating the hypothesis of shape similarity between soil particle-size distribution and water retention function","authors":"Ugo Lazzaro, Caterina Mazzitelli, Benedetto Sica, Paola Di Fiore, Nunzio Romano, Paolo Nasta","doi":"10.4081/jae.2023.1542","DOIUrl":"https://doi.org/10.4081/jae.2023.1542","url":null,"abstract":"Two pedotransfer functions (PTFs) are available in the literature enabling the soil water retention function (WRF) to be estimated from knowledge of the soil particle-size distribution (PSD), oven-dry soil bulk density (b), and saturated soil water content (s): i) the Arya and Heitman model (PTF-AH) and ii) the Mohammadi and Vanclooster model (PTF-MV). These physico-empirical PTFs rely on the hypothesis of shape similarity between PSD and WRF, and do not require the calibration of the input parameters. In the first stage, twenty-seven PSD models were evaluated using 4,128 soil samples collected in Campania (southern Italy). These models were ranked according to the root mean square residuals (RMSR), corrected Akaike Information Criterion (AICc), and adjusted coefficient of determination (R2adj). In the second stage, three subsets of PSD and WRF data (DS-1, DS-2, and DS-3), comprising 282 soil samples, were used to evaluate the two PTFs using the best three PSD models selected in the first stage. The hypothesis of shape similarity was assumed as acceptable only when the RMSR value was lower than the field standard deviation of the WRFs (*), which is viewed as a tolerance threshold and computed from the physically-based scaling approach proposed by Kosugi and Hopmans (1998). In the first study area (DS-1), characterized by a fairly uniform, loamy textured volcanic soil, the PTF-AH outperformed the PTF-MV and both PTFs provided reasonable performance within the acceptance threshold (i.e., RMSR < *). In the other two heterogeneous field sites (DS-2 and DS-3, characterized by soil textural classes that span from clay and clay-loam to loam and even sandy-loam soils), the PTF-MV (with 3% to 6% RMSR surpassing *) outperformed the PTF-AH (with 8% to 30% RMSR surpassing *) and the majority of RMSR values were larger than those obtained in the original studies. The mean relative error (MRE) revealed that the PTF-MV systematically underestimates the measured WRFs, whereas the PTF-AH provided negative MRE values indicating an overall overestimation. The outcomes of our study provide a critical evaluation when using calibration-free PTFs to predict WRFs over large areas.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"39 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134908214","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Marco Bovo, Shahad Al-Rikabi, Enrica Santolini, Beatrice Pulvirenti, Alberto Barbaresi, Daniele Torreggiani, Patrizia Tassinari
Controlling the microclimate condition inside a greenhouse is very important to ensure the best indoor conditions for both crop growth and crop production. To this regard, this paper provides the results of a novel approach to study a greenhouse, aiming to define a porous media model simulating the crop presence. As first, an experimental campaign has been carried out to evaluate air temperature and air velocity distributions in a naturally ventilated greenhouse with sweet pepper plants cultivated in pots. Then, the main aspects of energy balance, in terms of mass transfer and heat exchange, and both indoor and outdoor climate conditions have been combined to set up a computational fluid dynamics model. In the model, in order to simulate the crop presence and its effects, an isotropic porous medium following Darcy’s law has been defined based on the physical characteristics of the crops. The results show that the porous medium model could accurately simulate the heat and mass transfer between crops, air, and soil. Moreover, the adoption of this model helps to clarify the mechanism of thermal exchanges between crop and indoor microclimate and allows to assess in more realistic ways the microclimate conditions close to the crops.
{"title":"Definition of thermal comfort of crops within naturally ventilated greenhouses","authors":"Marco Bovo, Shahad Al-Rikabi, Enrica Santolini, Beatrice Pulvirenti, Alberto Barbaresi, Daniele Torreggiani, Patrizia Tassinari","doi":"10.4081/jae.2023.1540","DOIUrl":"https://doi.org/10.4081/jae.2023.1540","url":null,"abstract":"Controlling the microclimate condition inside a greenhouse is very important to ensure the best indoor conditions for both crop growth and crop production. To this regard, this paper provides the results of a novel approach to study a greenhouse, aiming to define a porous media model simulating the crop presence. As first, an experimental campaign has been carried out to evaluate air temperature and air velocity distributions in a naturally ventilated greenhouse with sweet pepper plants cultivated in pots. Then, the main aspects of energy balance, in terms of mass transfer and heat exchange, and both indoor and outdoor climate conditions have been combined to set up a computational fluid dynamics model. In the model, in order to simulate the crop presence and its effects, an isotropic porous medium following Darcy’s law has been defined based on the physical characteristics of the crops. The results show that the porous medium model could accurately simulate the heat and mass transfer between crops, air, and soil. Moreover, the adoption of this model helps to clarify the mechanism of thermal exchanges between crop and indoor microclimate and allows to assess in more realistic ways the microclimate conditions close to the crops.","PeriodicalId":48507,"journal":{"name":"Journal of Agricultural Engineering","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135216262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：