查看更多>>摘要：? 2022 Elsevier B.V.The lactation period is the first stage in a piglet's life cycle, and a piglet's activity during this time is an important indicator of the growth stage, better activity curve reflecting higher growth state. Detecting changes in physical activity may help in the early detection of unhealthy conditions and help a breeder carry out targeted treatment. However, both light interferences caused by devices such as heat preservation lamps and the difficulty of identifying piglets pose significant challenges for automated monitoring of piglet activity. In this study, we proposed an automated method for monitoring piglet activity. This method, named Frame Differences in combination with Convolution Neural Network (FD-CNN), was used to detect the regions of the active piglets by combining the frame difference method with the YOLOv5s network model. To estimate the whole average activity of piglets during lactation, the ratio of the area with active piglets to the area of all piglets was determined. The changing activity of day-old piglets was analyzed, and their status under different light source conditions was also evaluated. Video traceability was performed to detect the abnormal activity points, and the causes of these anomalies were studied. The results showed that our method could detect motion piglets (precision = 0.936) and quantify the activity status of piglets. When the detection frequencies were 6 s and 1 h, the similarity of activity value between the FD-CNN and manual detection were 58.36% and 78.90%, respectively. After the conversion of the activity metrics to values ranging from 0 to 1, the average activity value of day-old piglets with light was higher than 0.25 and that under the no-light condition was lower than 0.2. The traceability results of the abnormal activity points (activity value above 1) show that excess activity was mainly caused by sows attacking piglets or bumping into limits. The FD-CNN could replace the manual detection of piglet activity during lactation to a certain extent. The backtracking of abnormal activity points is beneficial for the timely detection of abnormal interaction behaviors between sows and piglets and provides technical support for the intelligent development of animal husbandry.
查看更多>>摘要：? 2022 Elsevier B.V.A serious outbreak of leaf-eating insects namely bagworm (Lepidoptera: Psychidae), especially Metisa plana species, may cause a 43% yield loss in oil palm production due to late proper control of bagworm populations. Identification of the bagworm instar stage is important to ensure proper control measures are applied in the infested area. This study aims to distinguish the bagworm larvae from second (S2) to fifth (S5) instar stages using hyperspectral imaging and machine learning technique. The capability of spectral reflectance and morphological features namely area, perimeter, major axis length, and minor axis length to classify the instar stage were studied. A total of 2000 sample points of larva were extracted from hyperspectral images. It was then followed by the identification of sensitive wavelengths of each stage using analysis of variance (ANOVA). Results show that seven wavelengths from the blue and green band (i.e., 470 nm, 490 nm, 502 nm, 506 nm, 526 nm, 538 nm, and 554 nm) gave the most significant difference in distinguishing the larval instar stages. To provide a more economical approach, only two wavelengths were used for model development. Later, the classifications models were developed separately using five different types of datasets: (A) significant morphological feature, (B) all significant wavelengths, (C) two wavelengths from the same spectral region, (D) two wavelengths from different spectral regions, and (E) two significant wavelengths and a significant morphological feature. Results have shown the dataset which used green bands at 506 nm and 538 nm with a weighted k-nearest neighbour classifier achieved the best value of accuracy (91% – 95%), precision (0.83 – 0.87), sensitivity (0.77 – 0.99), specificity (0.94 – 0.96) and F1-score (0.81 – 0.91). It was mainly due to green pigments which strongly correlates with the chlorophyll content of the frond leaves fed by the larvae to build and enlarge the case. The capability of the model to detect the young larval instar stages (S2 - S3) where an active feeding activity takes place allows quick decisions about outbreak control measures.
Villar L.B.Brignole N.B.De Meio Reggiani M.C.Vigier H.P....
15页
查看更多>>摘要：? 2022 Elsevier B.V.Evolutionary algorithms can efficiently be applied in broad practical issues by tailoring their operators to the specific combinatorial optimization problem under study. Based on Genetic Algorithms, this paper proposes a master–slave strategy enhanced with an ad hoc chromosome redefinition for the beekeeping value chain problem. Boolean and real genes are combined in a single chromosome to include a wide variety of decision-making optimization variables. Since production usually involves temporal dependency among processes, crossover and mutation operators were properly adjusted. Moreover, novel crossover and mutation operators were designed to make sure that all ensuing individuals were feasible. Since the evaluation of chromosomes may become time-consuming, parallel programming was adopted so that the Workers can simultaneously explore different instances. In particular, the model was implemented in order to optimize the beekeeping value chain aiming at the maximization of the Net Present Value. The results show that the improved algorithm is useful to make economic-financial decisions concerning the beekeeping activity in the southwest of Buenos Aires (Argentina). The proposed approach manages to boost the space search, always yielding realistic scenarios.
查看更多>>摘要：? 2022 Elsevier B.V.Plant breeding is an effective guarantee of agriculture production and food security, whereas information technologies are effective ways to promote plant variety improvement. Although breeding information technologies provide a convenient and scientific means for breeding, it also leads to the problem of massive data processing with multidimensional breeding data in multiple granularities across generations. Therefore, decision support tools that help breeders extract relevant and valuable information are necessary and important. This paper introduces the decision support tools of the golden seed breeding cloud platform at three levels: (i) graphical user interfaces; (ii) statistical tools; and (iii) intelligent decision support tools. The advantages have effectively improved the breeding efficiency as well as the data management and application capabilities in more than 150 breeding institutions in China. The successful application of our system confirms the effectiveness of our work and provides data and support for future research.
查看更多>>摘要：? 2022 Elsevier B.V.In log trade, precise measuring log volume is directly related to the economic benefits. Usually, the length of the same batch of wood is relatively fixed, but the shape and size of log end are different. Therefore, the variability of the size, shape and density of log end, and the sophisticated scenarios lead to great challenge for log detection from bundled log ends images. In addition, the large number of logs in the bundled log ends image results in less pixels occupied by a single log, and it is essential to detect the small log ends. In order to accurately detect each log from bundled log ends image, a new approach of log detection based on K-median clustering and improved You Only Look Once (YOLO)v4-Tiny network has been developed. The specific methods are as follows: (1) In consideration of the characteristics of log size, shape and its distribution, the K-median clustering method is used to select a more appropriate size of multi-scale anchor boxes to achieve a more consistent detection boxes for the log ends; (2) The detection scales of YOLOv4-Tiny network are increased to three and the spatial pyramid pooling (SPP) module is added to enhance the ability of feature extraction for small targets, such as small log end; (3) The self-attention mechanism based on squeeze and excitation (SE) is inserted into the deep structure of the network, which automatically determines the threshold without relevant professional knowledge to eliminate the high-dimensional noise that may result in error gathering of the centers of log ends and predicted boxes. The Precision, Recall and F1-score of YOLOv4-Tiny in the experimental test set are 91.87%, 94.91% and 0.93 respectively, while the three indicators are improved to 93.97%, 95.34% and 0.95 respectively by using the proposed model. Moreover, the complete intersection over union (CIoU) loss of our model is 2.46 and is reduced by 51.48% compared to the YOLOv4-Tiny with 5.07, which means the predicted boxes of our model are closer to the target bounding boxes. Consequently, the experiment results demonstrate that the performance of the proposed approach is better than that of YOLOv4-Tiny network.
查看更多>>摘要：? 2022Breeding strategies for many crops require quantitative evaluations of many genotypes from within as large of a diverse breeding pool as possible. In selecting pathogen tolerant genotypes, accurate and fast phenotyping to investigate genetic responses to pathogen infection and reproduction is crucial. Analyzing leaf tissues with spectral tools along with ground truth data offers potentially large gains in screening efficiency. However, ground truth labels per plant may not capture the effects of asymptomatic leaves and heterogeneous canopy responses to stress. We explored a semi-supervised clustering-based technique in which spectral patterns unique to and common among leaves from nematode infected plants are distinguished from patterns with no relationship to infection; we utilize these spectral patterns in ranking genotype tolerances to infection as a secondary objective. Proximal hyperspectral leaf scans (360 nm–1700 nm) of three walnut rootstock genotypes (MS1 122, VX211, MS1 127) were used in an agglomerative clustering procedure based on spectral angle mapper (SAM) distances to choose a spectral endmember representing the root lesion nematode, Pratylenchus vulnus, stress symptom on leaf per genotype. The histogram of SAM distances between control samples and the endmember was calculated. Next, the histogram of SAM distances between infected samples and the endmember was calculated. The shift between these histograms was then found using the minimum difference of pair assignments (MDPA) measure. The MDPA measures were 4.88, 3.14, and 5.48 for MS1 122, VX211, and MS1 127, respectively. This meant a genotype ranking in the order VX211, MS1 122 and, MS1 127 from the least affected by nematode infection to most impacted, which agreed with classification by nematological examinations of the plants. Clustering leaves based on their spectral response has the potential to overcome the limitation of heterogeneous canopy responses to stress in high-throughput phenotyping and other applications.
查看更多>>摘要：? 2022 The AuthorsThis study introduces a new approach combining a mechanistic greenhouse climate model and a disease model for the forecast of diseases occurrence in greenhouses. The method was evaluated in NPADB (National Precision Agriculture Demonstration Base), Beijing, China using data collected from transplanting to the primary infection that occurred in the greenhouse, in the spring season of 2021. First, the dynamic model is used to predict the greenhouse indoor climate 72 h ahead. Then, this prediction is used as input to the disease model to detect disease occurrence in advance. The predictions for the greenhouse downy mildew were compared using real-time measured data for two months. After several false-positive reports, one positive report by both methods fitted the first observation in the greenhouse on April 24, 2021. Thus, the main contribution of this work is the early warning cucumber downy mildew via coupling climate and disease models, where only transient inputs from weather forecasts are required.
查看更多>>摘要：? 2022 Elsevier B.V.In the remote sensing discrimination of tree species, the use of multi-temporal images may considerably improve the accuracy of recognising tree species. This study aimed to explore the effectiveness of the RedEdge-MX sensor, which was used to capture images during the key node period of tree growth, for driving the identification of tree species. The spectral band, texture and digital surface model (DSM) features of these data and their combinations were used as data sets, and 32 tree species were classified using maximum likelihood classification and random forest. The results demonstrated that the image imaging period considerably influenced the recognition of tree species. The recognition accuracy of tree flowering and leafing period data was the highest (52.98%, 86.66% and 86.90% based on spectral, texture and spectral + texture + DSM features, respectively), whereas that of the leafy period data was the lowest (34.32%, 82.39% and 82.81%). The classification accuracy significantly improved (72.76%, 91.51% and 92.16%) when multi-temporal data were combined. Moreover, the texture extraction window significantly affected the classification accuracy. The accuracy was low (70.47%, four seasons) when the minimum window was used, and this rate increased to 91.52% when the appropriate window was used but declined when an excessively large window was used. Furthermore, the classification accuracy obtained using texture features was higher than that of the spectral bands and DSMs were used. Their combination optimised the accuracy of tree species classification (92.16%). These results show that the capture of key node images in tree growth, the selection of appropriate texture extraction windows, and the use of multiple types of features can drive the effective recognition of urban greening tree species.
do Nascimento F.A.L.da Silva A.J.P.Freitas F.T.O.Fernandes R.D.M....
11页
查看更多>>摘要：? 2022 Elsevier B.V.Principles and frequency of soil water content (SWC) measurements used in the transient water flow experiments may cause distinction on the determination of soil hydraulic properties (SHP) by inverse modelling (IM). Consequently, use of agro-hydrological models and irrigation management can be affected. This work has the following objectives: (i) to analyze if the type of sensor and the frequency of SWC measurement used in the transient water flow experiments for IM affect the results regarding SHP; (ii) to assess if variations on the SWC measurement frequencies affect the results of SHP to the point of causing effects on the agro-hydrological modelling through SWAP model and on the passion fruit irrigation management based on soil water sensing; (iii) point out strategies for the correct irrigation management based on SHP obtained by IM. The results regarding SHP were applied to the modelling of crop evapotranspiration (ETc) and SWC with the use of SWAP agro-hydrological model. The type of sensor and the frequency of SWC data measurements interfered in the results regarding SHP obtained by IM, with implications on the agro-hydrological modelling and on the irrigation management through soil water sensing (SWS). The results of this study alert for the differences that may occur in SWC values when those are obtained from matric potential through soil water retention curves. This is specially important for higher and lower critical matric potentials used as reference for the moment to turn on and off a irrigation. When using GS1 sensors for IM, it is recommended to use intervals of SWC data measurement of 1 h, as the TDR sensors performed better with SWC data acquisition frequency between 1 and 12 h. Finally, it is understood that the factors that affect agro-hydrological modeling and the performance of the type of soil water sensor are diverse and integrated in a complex nonlinear way, for instance: atmospheric condition, soil texture and structure, topography, presence/absence of plants, soil and irrigation management practices. Thus, similar studies should be carried out considering variations in these factors so that a generalist and integrated conclusion is possible.
查看更多>>摘要：? 2022 Elsevier B.V.3D point cloud reconstruction, as the key technology to obtain high-throughput fruit phenotypic data, has solved the problems caused by complex environments, high fruit similarity, and the lack of public datasets suitable for fruit characterization. However, in the process of identifying and segmenting fruit data from point cloud, the existing network architectures lead to problems such as classification error, incomplete segmentation and low efficiency. In this paper, we introduce LFPNet, a novel and efficient lightweight neural network that directly consumes fruit point clouds in the real scene. Our network mainly has the following three advantages: 1) The introduction of voxel-filter based down-sampling preprocessing can help to avoid classification error caused by invalid noise interference. 2) A 3D STN is designed to solve the lack of spatial invariance in convolutional neural network (CNN) when calculating and analyzing fruit point clouds. 3) By introducing spatial pyramid pooling and combining local and global features, a fruit segmentation network is built to improve the integrity of segmentation in fruit scenes. Experimental results show that our LFPNet performs as well as or better than most of its peers in terms of classification accuracy and segmentation integrity.
NSTL
Elsevier