Computers and Electronics in Agriculture最新文献

{"title":"Mining sensitive hyperspectral feature to non-destructively monitor biomass and nitrogen accumulation status of tea plant throughout the whole year","authors":"","doi":"10.1016/j.compag.2024.109358","DOIUrl":"10.1016/j.compag.2024.109358","url":null,"abstract":"<div><p>Rapid and non-destructive estimation of tea plant growth and nitrogen (N) nutrition status using hyperspectral remote sensing is crucial for precise management of tea gardens. This study aimed to mine and fuse sensitive hyperspectral features to achieve an accurate estimation of tea plant growth parameters (biomass and N accumulation) throughout the whole year. An ASD Handheld 2 sensor was used to collect canopy hyperspectral reflectance of tea plants across four periods (Period 1–4) within a year, with tea plant biomass and N accumulation indicators acquired synchronously. The measured spectral reflectance and its first derivative, and wavelet feature were extracted and used to establish quantitative relationships with tea plant growth parameters. Random forest and LASSO algorithms were employed to combine sensitive hyperspectral features and construct the biomass and N accumulation monitoring models. The results showed that wavelet features (R² = 0.35–0.58) had a stronger correlation with tea plant biomass and N accumulation parameters compared with the measured reflectance or first derivative spectral features. Similarly, the hyperspectral indices (R² = 0.51–0.69) derived from sensitive wavelet features performed an accurate estimation of tea plant growth parameters. Furthermore, the combination of sensitive hyperspectral indices derived from measured reflectance, first derivative, and wavelet feature using random forest (R² = 0.67–0.76) and LASSO (R² = 0.61–0.72) algorithms achieved the greatest accuracy for monitoring tea plant biomass and N accumulation compared with individual hyperspectral feature. Additionally, the above estimation models obtained higher accuracy in period 4 compared to periods 1–3. This study provides valuable remote sensing technical support for predicting biomass and N accumulation status of tea plant throughout the whole year.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142021326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quality monitoring of glutinous rice processing from drying to extended storage using hyperspectral imaging","authors":"","doi":"10.1016/j.compag.2024.109348","DOIUrl":"10.1016/j.compag.2024.109348","url":null,"abstract":"<div><p>The quality of glutinous rice (GR) is susceptible to deterioration and losses due to biological or environmental factors during storage. Traditional quality assessment techniques are often time-consuming and challenging. In this study, a rapid and reliable hyperspectral imaging (HSI) technique is utilized to monitor GR quality during storage. Paddy samples were dried at 50 °C, 60 °C and 70 °C. Subsequently, these samples were milled and stored under three conditions: freeze storage (−10 °C), cold room (6 °C) and ambient (∼26 °C) for 6 months. The methodology involved data acquisition from both HSI and standard references methods, with data on hyperspectral reflectance, head rice yield (HRY), broken rice yield (BRY) and milled rice yield (MRY) collected every two weeks. Five machine learning (ML) models were evaluated for quality prediction using Python3, with Random Forest (RF) identified as the best performer, achieving a coefficient of determination (R²) of 0.995. Hyperparameter tuning (HPT) further improved the RF model’s R² by 0.3 %. Parity plot analysis confirmed the accuracy of the RF model in describing GR quality during storage. The study demonstrates the significant impacts of different storage and drying temperatures on HSI data and GR quality attributes. Significant differences in reflectance were observed, with higher reflectance for samples dried at 60 °C and freeze-storage, while lower reflectance for samples dried at 70 °C and cold-room storage. These findings align with reference method results and ML predictions, revealing that drying paddy at 60 °C and storing it under freeze conditions enhances HRY and increases the commercial value of GR. Overall, this study highlights the potential of the HSI for real-time quality monitoring of GR and its applicability to other grains.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142021328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient occlusion avoidance based on active deep sensing for harvesting robots","authors":"","doi":"10.1016/j.compag.2024.109360","DOIUrl":"10.1016/j.compag.2024.109360","url":null,"abstract":"<div><p>With the increasing shortage of agricultural labor, the development of harvesting robots is becoming more and more urgent. Most of them require vision to locate the target, however, occlusion is common in agricultural environment, which restricts the accuracy of visual target recognition, and even leads to failure in serious cases. The active perception method is an effective means, but how to efficiently find the best observation position remains difficult to avoid the waste of time caused by repeated invalid motion. Targeting these problems, an active deep sensing method is proposed for harvesting clustered and single fruits. First, the region of interest of the target is extracted by a segmentation network, and then the occlusion status of it is obtained by image processing methods. Taking the current observation position as the starting point, the camera is moved within a matrix to form confidence and occlusion rate distribution maps. After establishing a series of occlusion rate and confidence matrix datasets, a designed deep network has been trained, which is used to predict the maximum confidence/minimum occlusion rate position after the current occlusion status is estimated. To verify the reliability of the method, laboratory and field experiments were carried out for apples and clustered tomatoes. After 1000 times of verification, results show that the successful pick/recognition rate is increased by 38.7 %, and the average successful recognition time is 5.2 s, which is 63.1 % and 46.4 % faster than that of a fixed movement method and a simple heuristic method.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142021327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Measuring ornamental tree canopy attributes for precision spraying using drone technology and self-supervised segmentation","authors":"","doi":"10.1016/j.compag.2024.109359","DOIUrl":"10.1016/j.compag.2024.109359","url":null,"abstract":"<div><p>Tree canopy attributes or characteristics, such as canopy volume and density, are important parameters for calculating the precise quantity of agrochemicals required in each tree. Assessing the agrochemical needs in a nursery can help growers effectively estimate expenses, mitigate the risks of overuse or underuse, enhance resource utilization, and promote sustainable agricultural practices. This study aimed to develop an unmanned aerial vehicle (UAV)-based system to measure tree canopy attributes like tree height, tree count, canopy area and canopy volume using an image processing and self-supervised zero-shot segmentation approach. A high-resolution red-green-blue (RGB) sensor mounted on a drone captured three aerial image datasets, D1 and D2 from the first experimental plot and D3 from the second experimental plot. The acquired aerial images were stitched together and processed to generate a digital surface model (DSM) and a digital terrain model (DTM). A self-supervised zero-shot segmentation model, Segment Anything Model (SAM), was applied for tree segmentation and counting. The tree canopy area was calculated by segmenting individual trees and applying a conversion factor to convert pixelwise areas into square meters. Height calculation was done using elevation data of each pixel from DSM-DTM imagery within SAM-derived bounding box coordinates, and the highest pixel was considered as the height of the tree. The drone-calculated heights of 24 randomly selected trees were compared with manually measured heights in all datasets. The results showed an average absolute error of 2.43% in D1, 7.09% in D2 and 12.58% in D3. The Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) were noted to be 0.09 m and 0.23 m in D1, 0.13 m and 0.25 m in D2, and 0.16 m and 0.21 m in D3, respectively. In D3, manual area and volume calculations were performed for validation. The RMSE and MAE for area calculation were calculated as 0.18 m² and 0.15 m², respectively, and the RMSE and MAE for volume measurements were calculated as 0.33 m³ and 0.26 m³ respectively, which indicated the level of agreement between manual and drone measurements. Using canopy volume information, calculated by summing pixel heights within each tree’s bounding box and multiplying by the ground sample distance (GSD) of 0.025 m per pixel, the average quantity of agrochemicals needed for a Maple tree was estimated at 0.20 L. These results underscore the high potential of this method for accurate canopy characteristic calculations and precision spray applications in the ornamental industry.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142040651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detection of kernels in maize forage using hyperspectral imaging","authors":"","doi":"10.1016/j.compag.2024.109336","DOIUrl":"10.1016/j.compag.2024.109336","url":null,"abstract":"<div><p>In this study, the use of hyperspectral imaging was examined to enhance the kernel processing assessment of maize forage, which is crucial for optimizing the production of dairy cow feed and biogas production. As the visual contrast between the kernels and other crop particles is quite limited, spectral imaging could provide better kernel classification. A pushbroom hyperspectral imaging system was used, scanning samples collected during maize harvesting in the 400–1000 nm range. A PLSDA model for pixel classification was developed to distinguish kernel spectra from the other particles in the forage, achieving a pixel-level classification accuracy of 95.2 %. Next, the most important wavelengths were identified by means of a stepwise procedure using the Wilks Lambda criterion. A Pixel classification using the top five discriminating wavelengths achieved nearly the same accuracy as the full spectrum wavelength model (95.2 % compared to 93.5 %) and did considerably better than the RGB classifier’s 86.3 % accuracy. Finally, these top 5 discriminating wavebands were applied in an object detection deep learning model, more specifically a Faster R-CNN model. While object detection based on these 5 wavebands stilled outperformed object detection based on RGB wavebands, detection performance, as measured by AP50, was rather low. This weak performance resulted from the low resolution of the hyperspectral imaging camera.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142012745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MAIANet: Signal modulation in cassava leaf disease classification","authors":"","doi":"10.1016/j.compag.2024.109351","DOIUrl":"10.1016/j.compag.2024.109351","url":null,"abstract":"<div><p>Cassava is the third largest source of carbohydrates for human consumption worldwide; however, it is highly susceptible to viral and bacterial diseases, which pose a significant threat to food security. The advancement of deep learning algorithms in precision agriculture holds the key to enabling the early classification of plant diseases, thereby leading to enhanced crop yields and ultimately stabilizing food security. In the coarse-grained label discrimination task of weak supervision learning, high-quality semantic features contain abundant semantic description information, which plays a crucial role in constructing a precise description of plant disease discrimination in tanglesome field circumstances and directly influences the performance of neural networks. Thus, a multiattention IBN anti-aliasing neural network (MAIANet) was proposed to improve the classification accuracy of cassava leaf disease classification by improving the feature quality in the coarseness label classification task. The proposed MAIANet neural network includes two innovative approaches. First, the multiattention method was designed to scale the feature signals twice to adjust the angular frequency of the feature signals in the residual branch for optimal feature fitting within the residual unit. Second, the anti-aliasing block extracts the high-frequency component feature and optimizes the quantization result of the pooling operation to depress the aliasing signal in the down-sampled feature maps. When the proposed method was tested and validated on the cassava dataset, the results showed that the prediction accuracy of the proposed method significantly improved, with an accuracy of 95.83 %, a loss of 1.720, and an F1-score of 0.9585, outperforming V2-ResNet-101, EfficientNet-B5, RepVGG-B3g4, and AlexNet with significant margins. Based on the above experimental results, the proposed algorithm is suitable for classifying cassava leaf diseases.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142012747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simulating droplet distribution characteristics in sprinkler irrigation using a modified ballistic model under multifactor coupling","authors":"","doi":"10.1016/j.compag.2024.109352","DOIUrl":"10.1016/j.compag.2024.109352","url":null,"abstract":"<div><p>External factors affecting the processes of sprinkler irrigation water flow generation, flight, and landing have not been thoroughly considered in existing ballistic models. This result indicates that ballistic models with better prediction effects under specific conditions are not sufficient for extension to multi-factor coupled scenarios in large-scale farmlands. Therefore, wind, evaporation, surface slope, and tilted sprinkler riser factors were comprehensively considered in this study. Differential equations for jet and droplet motion under the influence of wind, differential equations of droplet evaporation, sprinkler riser deflection angle matrix, and surface slope angle matrix were constructed to establish a droplet distribution model for sprinkler irrigation considering multifactor coupling using MATLAB 2018a software. The results showed that, under different working conditions, the data points of the droplet landing diameter, velocity, and angle were distributed near the 1:1 line. The Nash efficiency coefficients (NSE) for the droplet landing diameter, velocity, and angle varied from 0.821 to 0.932, 0.616 to 0.931, and 0.770 to 0.911, respectively. The increase in slope resulted in droplets with diameters larger than 4.63 mm concentrating on the land in the reverse slope direction. When the ambient temperature increases from 10 to 45 °C and the total evaporation rate increases from 0.45 to 4.37 %, the larger droplets have a larger area of contact with the air, and the higher the temperature, the greater the energy loss to the larger droplet diameters. The higher the wind speed, the more droplets in the downwind direction fall to the ground at a smaller landing angle, which can easily increase the risk of soil shear damage. If the sprinkler riser was tilted east, the droplets on both the east and west sides tended to be distributed centrally; the maximum droplet landing velocity occurred on the east side (tilted side), and the maximum droplet landing angle occurred on the west side. This study considers various factors that may affect the motion of sprinkler irrigation water flow, extends the application scenarios of the theoretical model, and improves the applicability of the theoretical model for sprinkler irrigation droplet motion in more complex and practical agricultural environments.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142012744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Green fruit detection methods: Innovative application of camouflage object detection and multilevel feature mining","authors":"","doi":"10.1016/j.compag.2024.109356","DOIUrl":"10.1016/j.compag.2024.109356","url":null,"abstract":"<div><p>Accurate detection of object fruits is essential for optimizing picking efficiency and predicting fruit yields. However, detecting early-stage ripening or green fruits, especially in complex fields, poses significant challenge due to their similarity to green leaves. This study introduces the TEAVit model, a novel camouflage object detection network specifically tailored for identifying green tomatoes in intricate agricultural environments. TEAVit incorporates a texture-edge-awareness module (TEAM) to enhance the extraction ability of texture feature by combining high-level and low-level features, an edge-guided feature module (EFM) to address background complexities and occlusions, and a context-aggregation module (CAM) to leverage contextual semantics. Experimental validation results demonstrate that the S-measure, E-measure, and F-measure performance metrics all exceed 90% on the tomato dataset, accompanied by a mean absolute error of 0.0245. These findings underpinned the effectiveness of the proposed green fruit camouflage object detection algorithm in offering new insights for agricultural target localization.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142012746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"End-to-end stereo matching network with two-stage partition filtering for full-resolution depth estimation and precise localization of kiwifruit for robotic harvesting","authors":"","doi":"10.1016/j.compag.2024.109333","DOIUrl":"10.1016/j.compag.2024.109333","url":null,"abstract":"<div><p>Full-resolution depth estimation within operational space of robotic arms and accurate localization of kiwifruits is very important for automated harvesting. Depth estimation is expected to be accurate and full-resolution while current depth estimation methods are susceptible to depth missing due to occlusion and uneven illumination. And depth estimation mostly focuses on fruit localization, while obstacles such as branches and wires, which can affect harvesting strategy, have not been considered. This paper localized kiwifruits based on bounding boxes output by YOLOv8m and full-resolution depth from an end-to-end stereo matching network, i.e., LaC-Gwc Net, which was trained after generating a stereo matching dataset by proposing a two-stage partition filtering algorithm. Results showed that LaC-Gwc Net achieved an end-point error (<em>EPE</em>) of 3.8 pixels, which means that accurate depth estimation can also be achieved for thin obstacles such as the branches and the wires. Additionally, YOLOv8m obtained acceptable results in detecting kiwifruits and their calyxes, reaching mean average precision (<em>mAP</em>) of 93.1% and detection speed of 7.0 ms. The methodology obtained only kiwifruit localization error of 4.0 mm on the Z-axis, which meets requirements of robotic harvesting. Furthermore, this study considered the localization of obstacles in kiwifruit orchards, providing high-precision full-resolution depth estimation for agricultural harvesting robots.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142012756","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantitative analysis and planting optimization of multi-genotype sugar beet plant types based on 3D plant architecture","authors":"","doi":"10.1016/j.compag.2024.109231","DOIUrl":"10.1016/j.compag.2024.109231","url":null,"abstract":"<div><p>The type of crops plays a critical role in determining the canopy light interception and is a decisive factor for yield. Thus, it is of significant importance to have a comprehensive understanding of the similarities and differences in plant type for crop improvement. In this study, the Structure-from-Motion in conjunction with multi-view stereo (SfM-MVS) method was employed to capture multi-angle images of 132 sugar beet varieties at two growth stages, from which three-dimensional(3D) point clouds were reconstructed for all individual sugar beets. Nine plant phenotypic traits were extracted based on the point clouds, and their correlations and heritability were calculated. An unsupervised machine learning approach was utilized to classify all varieties based on their plant type, and the characteristics of different types were statistically analyzed. Subsequently, a variety of different canopies were simulated, and a ray-tracing software was used to simulate light interception of the day. The results revealed that sugar beet plants could be roughly classified into five distinct types with significant differences of the structure. The coefficient of variation of phenotypic parameters for all varieties was 33.2 % in July and decreased to 26.7 % in August. The heritability similarly declined from 0.82 to 0.50, indicating that the structure of the sugar beet plants was exacerbated by environmental influences as the growing season progressed. The light interception results showed that intercropping with different plant types had different effects on light interception, with differences in light interception of up to 1000 W/h across the canopy in July, but this effect was not always favorable, and a decrease in the total amount of light interception also occurred in intercropping with different plant types compared to monocropping.</p></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142006392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}