Information Processing in Agriculture最新文献

Accurate and fast detection of abnormal hydroponic lettuce leaves is primary technology for robotic sorting. Yellow and rotten leaves are main types of abnormal leaves in hydroponic lettuce. This study aims to demonstrate a feasibility of detecting yellow and rotten leaves of hydroponic lettuce by machine learning models, i.e. Multiple Linear Regression (MLR), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM). One-way analysis of variance was applied to reduce RGB, HSV, and L*a*b* features number of hydroponic lettuce images. Image binarization, image mask, and image filling methods were employed to segment hydroponic lettuce from an image for models testing. Results showed that G, H, and a* were selected from RGB, HSV, and L*a*b* for training models. It took about 20.25 s to detect an image with 3 024 × 4 032 pixels by KNN, which was much longer than MLR (0.61 s) and SVM (1.98 s). MLR got detection accuracies of 89.48% and 99.29% for yellow and rotten leaves, respectively, while SVM reached 98.33% and 97.91%, respectively. SVM was more robust than MLR in detecting yellow and rotten leaves of hydroponic. Thus, it was possible for abnormal hydroponic lettuce leaves detection by machine learning methods.

Historic maps showing the temporal distribution of rice fields are important for precision agriculture, irrigation optimisation, forecasting crop yields, land use management and formulating policies. However, mapping rice fields using traditional ground surveys is impractical when high cost, time and labour requirements are considered, and the availability of such detailed records is limited. Although satellite remote sensing appears to be a viable solution, conventional segmentation and classification methods with spectral bands are often unable to contrast the distinct characteristics between rice fields and other vegetation classes. To this end, we explored a novel, Google Earth Engine (GEE) based multi-index random forest (RF) classification approach to map rice fields over two decades. Landsat images from 2000 to 2020 of two Sri Lankan rice cultivation districts were extracted from GEE and a multi-index RF classification algorithm was applied to distinguish the rice fields. The results showed above 80% accuracy for both training and validation, when compared against high spatial resolution Google Earth imagery. In essence, multi-index sampling and RF together synergised the compelling classification accuracy by effectively capturing vegetation, water (ponding) and soil characteristics unique to the rice fields using a single-click approach. The maps developed in this study were further compared against the MODIS land cover type product (MCD12Q1) and the corresponding superior statistics on rice fields demonstrated the robustness of the proposed approach. Future work seeking effective index combinations is recommended, and this approach can potentially be extended to other crop analyses elsewhere.

Accurate calculation for comprehensive power load of fishery energy internet plays a significant role in reasonable using of energy and reducing environmental pollution. However, as fishery power load is of greatly unique meteorology sensitivity, it continues to be a difficult problem. Therefore, the research of fishery meteorology is an important part of the rational development of fishery resources, the protection of production safety, and the pursuit of high and stable yield. This paper makes a deep study on the power load of the fishery energy internet under the influence of fishery meteorology and takes onshore fish pond as the research object. First of all, the power load is divided into three parts: oxygen enrichment power load, feeding power load, and water replenishment and drainage power load. The impact mechanism of fishery meteorology (including temperature, surface wind speed, precipitation, relative humidity, etc.) on it is described, and then the overall power load is obtained through modeling and integration. Finally, taking the Yuguang Complementary Project in Zhouquan Town, Tongxiang, Zhejiang Province, China as an example, using the meteorological data of its typical spring day and using the MATLAB tool to solve, the hourly comparison of the three types of power loads, the comprehensive power load demand, the full-day electricity charge forecast and the total annual power consumption are calculated. The annual power consumption per hectare and per kilogram of output calculated by simulation are basically consistent with the order of magnitude of the survey data, which proves the validity of the model proposed. The model established in this paper is an original work, and the exploration of fishery energy internet can draw lessons from it.

In unstructured environments, dense grape fruit growth and the presence of occlusion cause difficult recognition problems, which will seriously affect the performance of grape picking robots. To address these problems, this study improves the YOLOX-Tiny model and proposes a new grape detection model, YOLOX-RA, which can quickly and accurately identify densely growing and occluded grape bunches. The proposed YOLOX-RA model uses a 3 × 3 convolutional layer with a step size of 2 to replace the focal layer to reduce the computational burden. The CBS layer in the ResBlock_Body module of the second, third, and fourth layers of the backbone layer is removed, and the CSPLayer module is replaced by the ResBlock-M module to speed up the detection. An auxiliary network (AlNet) with the remaining network blocks was added after the ResBlock-M module to improve the detection accuracy. Two depth-separable convolutions (DSC) are used in the neck module layer to replace the normal convolution to reduce the computational cost. We evaluated the detection performance of SSD, YOLOv4 SSD, YOLOv4-Tiny, YOLO-Grape, YOLOv5-X, YOLOX-Tiny, and YOLOX-RA on a grape test set. The results show that the YOLOX-RA model has the best detection performance, achieving 88.75 % mAP, a recognition speed of 84.88 FPS, and model size of 17.53 MB. It can accurately detect densely grown and shaded grape bunches, which can effectively improve the performance of the grape picking robot.

The crop pests and diseases in agriculture is one of the most important reason for the reduction of bulk grain and oil crops and the decline of fruit and vegetable crop quality, which threaten macroeconomic stability and sustainable development. However, the recognition method based on manual and instruments has been unable to meet the needs of scientific research and production due to its strong subjectivity and low efficiency. The recognition method based on pattern recognition and deep learning can automatically fit image features, and use features to classify and predict images. This study introduced the improved Vision Transformer (ViT) method for crop pest image recognition. Among them, the region with the most obvious features can be effectively selected by block partition. The self-attention mechanism of the transformer can better excavate the special solution that is not an obvious lesion area. In the experiment, data with 7 classes of examples are used for verification. It can be illustrated from results that this method has high accuracy and can give full play to the advantages of image processing and recognition technology, accurately judge the crop diseases and pests category, provide method reference for agricultural diseases and pests identification research, and further optimize the crop diseases and pests control work for agricultural workers in need.

Crop height measurement is widely used to analyze and estimate the overall crop condition and the amount of biomass production. Not only is manual measurement on a large scale time-consuming but also it is not practical. Besides, advanced equipment is available but they require technical skills and are not reasonable for smallholders. This article investigates the feasibility of a simple and low-cost measurement system that can monitor crops height of paddy rice and wheat using laser technology. After designing and fabricating, this system was tested and evaluated in both laboratory and farm sections. In the laboratory, paddy rice height was measured, and in the field section, the height detection system measured wheat height. The results showed that the coefficient of determination (R²) between manual measurement and height detection system measurement for paddy rice was 0.96 and for wheat was 0.85. Besides, there was no significant difference between the two datasets at the level of 5%. Hence, this system can be a useful and accurate tool to monitor crops height in different growing steps.

Combining multiple crop protection Unmanned Aerial Vehicles (UAVs) as a team for a scheduled spraying mission over farmland now is a common way to significantly increase efficiency. However, given some issues such as different configurations, irregular borders, and especially varying pesticide requirements, it is more important and more complex than other multi-Agent Systems (MASs) in common use. In this work, we focus on the mission arrangement of UAVs, which is the foundation of other high-level cooperations, systematically propose Efficiency-first Spraying Mission Arrangement Problem (ESMAP), and try to construct a united problem framework for the mission arrangement of crop protection UAVs. Besides, to characterise the differences in sub-areas, the varying pesticide requirement per unit is well considered based on Normalized Difference Vegetation Index (NDVI). Firstly, the mathematical model of multiple crop-protection UAVs is established and ESMAP is defined. Furthermore, an acquisition method of a farmland’s NDVI map is proposed, and the calculation method of pesticide volume based on NDVI is discussed. Secondly, an improved Genetic Algorithm (GA) is proposed to solve ESMAP, and a comparable combination algorithm is introduced. Numerical simulations for algorithm analysis are carried out within MATLAB, and it is determined that the proposed GA is more efficient and accurate than the latter. Finally, a mission arrangement tested with three UAVs was carried out to validate the effectiveness of the proposed GA in spraying operation. Test results illustrated that it performed well, which took only 90.6 % of the operation time taken by the combination algorithm.

Impact bruise damage and quality of ‘Gim Ju’ guava were investigated for different drop heights and number of drops using fractal image analysis. For the impact test, a stainless-steel metal ball (250 g) was dropped on fruit from three drop heights (0, 0.3, 0.6 m) either once or five times. Fruit quality was evaluated for impact energy, bruise area (BA), bruise volume (BV), bruise susceptibility, bruise score and pulp color (L*, a*, b* and C values). The fractal dimension (FD) value using fractal image analysis was analyzed at the bruise region. Results showed that five drops (0.3 m) with a high impact energy (3 678.75 J) and a single drop (0.6 m) with a low impact energy (1 471.50 J) exhibited no significant in BA, BV, bruise score as well as all color values (L*, a*, b* and C). While the FD value of a single drop from 0.6 m had a higher FD value than that of five drops from 0.3 m. It is indicated that FD exhibited a better performance to classify impact bruising level of guava than BA, BV and color parameters. The FD value gradually decreased with increase of storage time and bruise severity. The correlation coefficient (r) values of FD (r =  − 0.794 and − 0.745) between BA and BV were more significant than those L* (r =  − 0.660 and − 0.615) and a* (r = 0.579 and 0.473). The coefficient of determination (R²) of the polynomial equation in bruised fruit (R² = 0.85 to 0.99) was greater than the control (no bruise) (R² = 0.80). A higher R²_val (0.88 and 0.92) was exhibited at five drops. Interestingly, FD analysis showed greater potential than color measurement to assess bruise impact damage in guava.

Nitrate is the primary water-soluble macronutrient essential for plant growth that is converted from excess fish feeds, fish effluents, and degrading biomaterials on the aquaponic pond floor, and when aquacultural malpractices occur, large amounts of it retain in the water system causing increase rate in eutrophication and toxifies fish and aquaculture plants. Recent nitrate sensor prototypes still require performing the additional steps of water sample deionization and dilution and were constructed with expensive materials. In response to the challenge of sensor enhancement and aquaponic water quality monitoring, this study developed sensitive, repeatable, and reproducible screen-printed graphite electrodes on polyvinyl chloride and parchment paper substrates with silver as electrode material and 60:40 graphite powder:nail polish formulated conductive ink for electrical traces, integrated with 9-gene genetic expression model as a function of peak anodic current and electrochemical test time for nitrate concentration prediction that is embedded into low-power Arduino ESP32 for in situ nitrate sensing in aquaponic pond water. Five SPE electrical traces were designed on the two types of substrates. Scanning electron microscopy with energy dispersive X-ray confirmed the electrode surface morphology. Electrochemical cyclic voltammetry using 10 to 100 mg/L KNO₃ and water from three-depth regions of the actual pond established the electrochemical test time (10.5 s) and electrode potential (0.135 V) protocol necessary to produce peak current that corresponds to the strength of nitrate ions during redox. The findings from in situ testing revealed that the proposed sensors have strong linear predictions (R² = 0.968 MSE = 1.659 for nSPEv and R² = 0.966 MSE = 4.697 for nSPEp) in the range of 10 to 100 mg/L and best detection limit of 3.15 μg/L, which are comparable to other sensors of more complex construction. The developed three-electrode electrochemical nitrate sensor confirms that it is reliable for both biosensing in controlled solutions and in situ aquaponic pond water systems.

As an important research topic in recent years, semantic segmentation has been widely applied to image understanding problems in various fields. With the successful application of deep learning methods in machine vision, the superior performance has been transferred to agricultural image processing by combining them with traditional methods. Semantic segmentation methods have revolutionized the development of agricultural automation and are commonly used for crop cover and type analysis, pest and disease identification, etc. We first give a review of the recent advances in traditional and deep learning methods for semantic segmentation of agricultural images according to different segmentation principles. Then we introduce the traditional methods that can effectively utilize the original image information and the powerful performance of deep learning-based methods. Finally, we outline their applications in agricultural image segmentation. In our literature, we identify the challenges in agricultural image segmentation and summarize the innovative developments that address these challenges. The robustness of the existing segmentation methods for processing complex images still needs to be improved urgently, and their generalization abilities are also insufficient. In particular, the limited number of labeled samples is a roadblock to new developed deep learning methods for their training and evaluation. To this, segmentation methods that augment the dataset or incorporate multimodal information enable deep learning methods to further improve the segmentation capabilities. This review provides a reference for the application of image semantic segmentation in the field of agricultural informatization.