首页期刊导航|Computers and Electronics in Agriculture
期刊信息/Journal information
Computers and Electronics in Agriculture
Elsevier Science Publishers
Computers and Electronics in Agriculture

Elsevier Science Publishers

0168-1699

Computers and Electronics in Agriculture/Journal Computers and Electronics in AgricultureSCIEIISTP
正式出版
收录年代

    Research on virtual Ackerman steering model based navigation system for tracked vehicles

    Zhang L.Zhang R.Li L.Ding C....
    13页
    查看更多>>摘要：? 2021A state feedback navigation control system based on a virtual Ackerman steering model is proposed in this study to address the problems of the poor smoothness of steering control and low control accuracy of tracked vehicles. The proposed system uses a real-time dynamic positioning global navigation satellite system. Furthermore, it incorporates cubic spline interpolation to smoothen a predetermined path. In addition, a path-tracking control method based on pole assignment, virtual Ackerman steering control model of a single-cylinder diesel-engine-powered tracked vehicle, and forward-steering proportional control method based on the principle of pulse-width modulation were designed. Simulations and field experiments were performed to test the application of the proposed system. The experimental results indicated that the absolute average and standard deviation of the path waypoint interval decreased from 0.227 and 0.348 m to 0.018 and 0.015 m, respectively. In 5- and 6-m-radius-circular and straight-line path-tracking simulations, the average error in path tracking was less than 0.002 m. For path tracking in an orchard environment, the average tracking error and standard error deviation between tree rows were 0.051 and 0.084 m, respectively. These results indicate that the proposed control system enable significantly stable control of single-cylinder diesel-engine-powered tracked vehicles, and its control accuracy meets the operational objectives of tracked vehicles.
      原文链接:
    • NSTL
    • Elsevier

    Data fusion based wireless temperature monitoring system applied to intelligent greenhouse

    Xia S.Nan X.Cai X.Lu X....
    12页
    查看更多>>摘要：? 2021 Elsevier B.V.In intelligent greenhouses, wireless sensor networks with uneven temperature distribution and low collection efficiency may lead to poor monitoring effects in real time. To improve the performance of the temperature monitoring system in intelligent greenhouses, a real-time fusion strategy for a hierarchical wireless sensor network (WSN) is proposed. The designed WSN has three layers. In the bottom, sensors collect and preprocess the temperature data of the greenhouse by an improved unscented Kalman filter. In the middle layer, each cluster-head sensor, as a local fusion center, is used to fuse the data collected from the bottom sensors by a parallel inverse covariance intersection fusion algorithm. In the top, a global fusion center is utilized to fuse the temperature data from the middle layer to reflect the global temperature of the greenhouse by an improved extreme learning machine algorithm. The designed algorithm applied in each layer ensures the efficiency and accuracy of data fusion in real time. Simulation results show that the designed fusion strategy effectively improves the fusion accuracy and realizes the real-time fusion. The performance of the designed temperature monitoring system is greatly improved.
      原文链接:
    • NSTL
    • Elsevier

    Mixture-based weight learning improves the random forest method for hyperspectral estimation of soil total nitrogen

    Lin L.Liu X.
    7页
    查看更多>>摘要：? 2021 Elsevier B.V.Visible and near-infrared reflectance spectroscopy is a promising technique to estimate soil total nitrogen, but because of many uncontrolled variations like soil water, developing a high-accuracy soil total nitrogen model is still challenging. This study proposed a new methodology called mixture-based weight learning to reduce the problems stated above. This method improves the estimation accuracy of soil total nitrogen by mixing soil total nitrogen with soil water, then using the random forest method to model soil total nitrogen. A series of different mixtures of soil water and soil total nitrogen were made, and based on the visible and near-infrared spectra measured under laboratory conditions, the optimal model (j = 0.03) was constructed and used as the final mixture-based weight learning–random forest model, which produced better results (R2 of validation = 0.757; root mean square error of validation = 0.235 g/kg; mean relative error of validation = 10.0%; ratio of performance to interquartile range of validation = 2.419) than a random forest model. Our mixture-based weight learning method combined with the random forest method has great potential for the accurate remote retrieval of soil total nitrogen and enhances the available ways to estimate soil properties.
      原文链接:
    • NSTL
    • Elsevier

    Research on profiling tracking control optimization of orchard sprayer based on the phenotypic characteristics of tree crown

    Nan Y.Zhang H.Zheng J.Yang K....
    9页
    查看更多>>摘要：? 2021 Elsevier B.V.The performance of the profiling tracking control system was one of the important factors affecting the spray deposition and drift characteristics of the profiling variable spray. This paper proposed a dynamic threshold function to improve the CMAC-PID control algorithm to improve the real-time response speed and robustness of profiling tracking control system. The optimal parameters of dynamic threshold function was determined through a response surface experiment and optimal calculation. The continuous profiling tracking experiment results for outdoor tree canopy within 0–2.5 s showed that, compared with the CMAC-PID control algorithm, when the improved CMAC-PID control algorithm adopted by the profiling control system, the rise time was respectively shortened by 71.5%, 66.1% and 67.3%, the adjustment time was respectively shortened by 67.6%, 57.6% and 50.0%, and the overshoot was respectively 0.658%, 0.552% and 3.46% for the profiling angle position step response of the profiling mechanism module A_up, B and A_down. The continuous profiling tracking experiment results for outdoor tree canopy within 2.5–11.5 s showed that, compared with the CMAC-PID control algorithm, when the improved CMAC-PID control algorithm was adopted by the profiling control system, the profiling tracking response curves and the profiling target angle curves were closer to coincidence, and the average error of profiling tracking has been reduced by 35.9%, 57.4% and 38.1% respectively for the profiling mechanism module A_up, B and A_down. It showed that the profiling control system using the improved CMAC-PID algorithm had better response speed and profiling tracking performance.
      原文链接:
    • NSTL
    • Elsevier

    Forest pest identification based on a new dataset and convolutional neural network model with enhancement strategy

    Liu Y.Xu J.Kong X.Xie L....
    14页
    查看更多>>摘要：? 2021 Elsevier B.V.Due to the lack of samples, deep learning in forest pest identification is severely limited, and classification accuracy and generalization ability are insufficient. To address this issue, we have constructed a new dataset of forest pests containing 67,953 images, enhanced the dataset by Graph-based Visual Saliency, and combined transfer learning and fine-tune to build a twice transfer strategy in the convolutional neural networks (CNNs) for pest recognition. Based on the new dataset and developed model, a new platform for forest pest identification was finally built. Compared with prevalent models including Inception-V3, MobileNet-V2, ResNet-50-V2, InceptionResNet-V2, and Xception, our method improves the accuracy and generalization ability of classification by 6.2% and 7.0%, respectively. Meanwhile, class activation maps show that the model's focus on the target has also been increased by 9.0%. In general, the new proposed dataset and training strategy can greatly improve classification performance of CNNs, which may be helpful to the effective control on forest pests.
      原文链接:
    • NSTL
    • Elsevier

    Two-step ResUp&Down generative adversarial network to reconstruct multispectral image from aerial RGB image

    Zhang Y.Yang W.Zhang W.Yu J....
    16页
    查看更多>>摘要：? 2021 Elsevier B.V.Convolutional neural network has brought breakthroughs on multispectral image reconstruction research. Previous work has largely focused on reconstructing MSI using the R-G-B channels from the MSI as inputs of the model. However, it's image manipulation rather than practical use. In real application, to reconstruct multispectral image using images from RGB camera is a research that has hardly been studied. In this research, high resolution aerial RGB images are collected by drone with RGB camera and multispectral images are collected by drone with RedEdge-M multispectral Camera. Then a new two-step Generative Adversarial Network (GAN)-based reconstruction method was proposed as follows: At first, MSI and RGB images are carefully registered to make sure that pixels are one–one correspondent. Then two data sources are cropped to form dataset. After that, a novel R-MSI GAN using is proposed. It uses a ResUp&Down block to replace the ResNet block of the Generator network and it outperforms ResNet-based GAN. The experimental results show that: (1) the combination of Mean Square Error and Discriminator (MSE-D) can alleviate the problem of the high-frequency loss of generated images. (2) The root means square error (RMSE), mean relative absolute error (MRAE) and Structural Similarity (SSIM) can only reflect overall error but can't reflect details in reconstructed images, while different bands' statistical histogram can present the total high-frequency loss of generated bands. (3) 3 indexes, which are intersection over union (IoU) based normalized difference vegetation index (NDVI)-IoU, normalized difference red edge (NDRE)-IoU and enhance vegetation index (EVI)-IoU, were defined to verify the effect of the generated MSI and they show good consistence with vegetation index map. 4 In comparisons among ResNet-based GAN, single step ResUp&Down GAN and two-step ResUp&Down GAN(T-GAN) with 3 loss functions (L1, MSE, Discriminator), the two-step ResUp&Down GAN(T-GAN) with MSE-D loss function performs best in reconstructing RGB bands. The T-GAN with L1loss-D (mean absolute error loss) performs best in reconstructing NIR and rededge bands. In summary, the proposed methods can effectively reconstruct MSI using images from RGB camera at drone based remote sensing.
      原文链接:
    • NSTL
    • Elsevier

    Comparison of the automated monitoring of the sow activity in farrowing pens using video and accelerometer data

    Oczak M.Bayer F.Vetter S.Maschat K....
    13页
    查看更多>>摘要：? 2021 The AuthorsPatterns in pigs activity can be an indicator of health and welfare of the animals. This motivates researchers to develop Precision Livestock Farming (PLF) tools for automated monitoring of pig activity level. In this research we compared two important technologies that can be used for this purpose, ear tag accelerometer and computer vision. Additionally, we compared both technologies with gold standard based on human labelling. A state-of-the-art object detection algorithm RetinaNet was trained on 9969 images and validated on 4273 images to automatically detect head of a sow, body of a sow, left ear, right ear and a hay rack. It was possible to detect these objects with a performance of 0.26 mAP@0.5:0.95. Activity of 6 sows was derived from detected parts of animals’ bodies and compared with activity measurement based on ear tag accelerometer data. Dynamic relation between activity measurement based on both technologies was modelled with Transfer Function (TF) models. For all 6 animals activity of the body of a sow based on object detection was very similar to accelerometer based activity measurement (R2 > 0.7). Similarly R2 between activity of a head of a sow and accelerometer based activity was also very similar for most sows (R2 > 0.7). Results of fitting of TF models to animal activity data based on ear tag accelerometer and output of object detection on body of sows and head of sows suggests that both technologies, the accelerometer and computer vision provide very similar information on activity level of animals. The presented computer vision method is limited to monitoring one animal under camera view as detected body parts cannot be associated with multiple individuals. Moreover, we expect that the method requires re-training the RetinaNet object detection algorithm with additional images collected on additional farms to achieve satisfactory performance in different environments. Application of computer vision approach might be advantageous in some PLF applications as it is non-invasive and might be less laborious than method based on ear tag accelerometer data.
      原文链接:
    • NSTL
    • Elsevier

    Machine learning imputation of missing Mesonet temperature observations

    Boomgard-Zagrodnik J.P.Brown D.J.
    11页
    查看更多>>摘要：? 2021Uninterrupted and reliable weather data is a necessary foundation for agricultural decision making, required for models based on accumulated growing degree days (GDD), chill units, and evapotranspiration. When a weather station experiences a mechanical or communications failure, a replacement (imputed) value should be substituted for any missing data. This study introduces a machine learning, network-based approach to imputing missing 15-minute and daily maximum/minimum air temperature observations from 8.5 years of air temperature, relative humidity, wind, and solar radiation observations at 134 AgWeatherNet (AWN) stations in Washington State. A random forest imputation model trained on temperature and humidity observations from the full network predicted 15-minute, daily maximum, and daily minimum temperature values with mean absolute errors of 0.43 °C, 0.53 °C, and 0.70 °C, respectively. Sensitivity experiments determined that imputation skill was related a number of external factors including volume and type of training data, proximity of surrounding stations, and regional topography. In particular, nocturnal cold air flows in the upper Yakima Valley of south-central Washington caused temperature to be less correlated with surrounding stations in the overnight hours. In a separate experiment, the imputation model was used to predict base- 10 °C GDD on 2020 July 1 trained entirely on 15-minute station data from previous years. Even with the entire season of observations missing, the model predicted the GDD value within an average error 1.4% with 125 of 134 stations within 5% of observations. Since missing data can typically be resolved within a timeframe of a few days, the network-based imputation model is a sufficient substitute for short periods of missing observational weather data. Other potential applications of an imputation model are briefly discussed.
      原文链接:
    • NSTL
    • Elsevier

    Developing machine learning models with multisource inputs for improved land surface soil moisture in China

    Wang L.Fang S.Zhuo W.Pei Z....
    11页
    查看更多>>摘要：? 2021 Elsevier B.V.Accurate and spatially continuous land surface soil moisture (SSM) data will greatly benefit analyses of heat transfer, energy exchange and agricultural dryness. To obtain spatiotemporally consistent SSM information, five machine learning (ML) models, i.e., polynomial regression (PR), ridge regression (RR), lasso regression (LR), elastic net regression (EnR) and random forest regression (RfR) models, were generated to map the regional SSM in the 0–10 cm soil layer across the study area. Multiple features, including the geographical location, elevation, vegetation coverage, soil texture, seasonal patterns and satellite-retrieved SSM product from Fengyun-3C (FY-3C), were selected as the input variables for the proposed ML models. In situ SSM measurements from the Chinese Automatic Soil Moisture Observation Stations (CASMOS) were used as the reference dataset. The error metrics, including the coefficient of correlation (R), mean relative error (MRE), unbiased RMSE (ubRMSE) and mean absolute error (MAE), between the measured SSM values and those estimated using the different models were calculated. Among those ML models, the RfR model showed the best performance in the training (R = 0.981, MRE = 7.3%, ubRMSE = 0.021 cm3/cm3, and MAE = 0.015 cm3/cm3) and testing (R = 0.789, MRE = 22.2%, ubRMSE = 0.065 cm3/cm3, and MAE = 0.047 cm3/cm3) processes and was applied to map the regional SSM values and measure the importance of each input feature. The results indicated that geographical location, i.e., latitude (35.84%) and longitude (16.96%), contributed the most to the SSM estimation model, followed by elevation (14.88%), vegetation coverage (9.75%), the FY-3C SSM product (8.30%), the soil texture (8.04%) and seasonal patterns (6.23%). In addition, the SSM estimations across mainland China matched the spatiotemporal patterns of historical precipitation well, which indicated the feasibility of achieving accurate and consistent land surface (0–10 cm) soil moisture monitoring results using the established RfR model with appropriately selected input features.
      原文链接:
    • NSTL
    • Elsevier

    Mastitis detection with recurrent neural networks in farms using automated milking systems

    Naqvi S.A.Deardon R.Barkema H.W.King M.T.M....
    10页
    查看更多>>摘要：? 2021 Elsevier B.V.Mastitis is the most important disease in the dairy industry. With widespread adoption of automated milking systems (AMS) in Canada, there is an increasing need for automated detection of mastitis in AMS farms. The main objective of this study was to develop a recurrent neural network (RNN) model for the detection of clinical mastitis (CM) in dairy cows on farms using AMS. Producer-recorded treatment records and AMS data were collected over 3 time periods from a total of 89 dairy farms in 7 provinces across Canada. In addition to developing effective models for the detection of CM, our study also evaluated different windows around the day of diagnosis when the cow would be considered CM-positive to guide practical implementation of models. We also compared numerous subsets of variables including milk and behavioural characteristics, cow traits and farm-level/environmental variables to determine their importance and impact on model performance. Data were randomly divided into a training and a hold-out test set, consisting of all records from 66 and 23 farms, respectively. A 10-fold internal cross-validation was also employed on the training set for model development. When comparing different windows of time around diagnosis, considering animals as CM-positive for 3 d prior to recorded diagnosis resulted in the most timely and effective detection of CM with a per-case sensitivity of 89.8% (range: 83.3–96.0%), and per-day specificity of 84.3% (range: 83.4–85.8%) over the validation folds. These levels of sensitivity and specificity were achieved when using all recorded variables and their daily variances, although the inclusion of behavioural variables and farm-level/environmental variables provided marginal performance improvement over using milk traits alone. Performance of the model was worse on the hold-out test set with a per-case sensitivity of 83.5% (range: 77.9–86.3%) and a per-day specificity of 80.4 % (range: 78.1–82.4%), likely due to farm-specific heterogeneity not encountered in the training data. Over 90% of cases of severe CM (defined by an increase in milk temperature over the pre-CM baseline) were identified by the model, indicating effective performance for the detection of CM requiring the most immediate treatment. Somatic cell count, daily variance in milking interval and milk temperature were identified as the 3 most important variables defined by their impact on model predictions. In addition to milking traits, 8 of the top 20 variables were behavioural measurements, suggesting they can play a role in the detection of CM. Daily variances also represented 8 of the 20 most important variables indicating that CM onset may be associated with sudden, within-day changes in the animal. In conclusion, this study demonstrates that RNNs are able to effectively detect CM by integrating a number of variables that are regularly measured on AMS farms but have typically been excluded from CM detection models.
      原文链接:
    • NSTL
    • Elsevier