查看更多>>摘要：? 2022Semen ziziphi spinosae are the seeds of sour dates. They can be used as both food and medicine, with large and full kernels producing particularly good medicinal herbs. However, the kernels of semen ziziphi spinosae are wrapped in a shell, so it is difficult to analyze the fullness of the kernel without damaging the seeds themselves. In this study, shortwave near-infrared (SW-NIR) and longwave near-infrared (LW-NIR) hyperspectral imaging techniques are used to determine the recognition ability of kernels from shell regions in different wavelength ranges using a two-band ratio algorithm. LW-NIR is found to be more suitable for the non-destructive detection of the inner kernel of semen ziziphi spinosae. Three practical wavelength selection algorithms and two classification algorithms are used to build pixel-level kernel detection models that achieve classification accuracies of greater than 97%. A novel improved Otsu algorithm is developed and applied to multispectral images. The watershed algorithm, threshold segmentation, and improved Otsu algorithm are used to build image-level kernel detection models. The detection results indicate that the improved Otsu algorithm is superior to other algorithms, achieving accuracy levels of 100%, 97.5%, 95%, and 100% for four different types of plumpness, respectively. These results indicate that the proposed method has the potential for the online detection of suitable semen ziziphi spinosae kernels. Moreover, LW-NIR hyperspectral transmission imaging combined with the improved Otsu algorithm is shown to provide a useful non-destructive method for detecting kernel fullness for medicinal materials.
查看更多>>摘要：? 2022Poultry farming is an essential industry of animal husbandry, which is developing in the direction of scale, intelligence and unmanned. Intelligent information perception ideas for different application scenarios are stimulated with the rapid development of sensors, information communication and robotics technologies and artificial intelligence-based information processing technologies worldwide. Intelligent perception of information in the poultry farming process is important for the liberation of labor, safeguarding animal welfare, and improving the automation and efficiency of poultry farming. In this review, the information perception framework in modern poultry farming has been analyzed to fully illustrate the important role of information perception in modern poultry farming. In addition, we have reviewed the research of information perception technology in poultry farming in 26 countries around the world. According to the different research goals, it is mainly divided into five aspects: individual poultry target perception, behavior recognition, health and environmental monitoring, and breeding-related robots. Some tables are provided to summarize and review research information on different topics. We found that for the information perception in poultry breeding, many challenges still need to be solved, such as the accurate perception of poultry individual information in complex environment, multi-scale monitoring of poultry house environment, intelligent disease diagnosis and exception handling, robot function expansion and multi-robot coordination. To achieve the goal of accurate, efficient, and intelligent perception of information in the unmanned poultry farming system, we have also pointed out the future research focus and development trends respectively by combining the characteristics of different problems.
查看更多>>摘要：? 2022 Elsevier B.V.Farm Management Information Systems (FMIS) have undergone tremendous development in the recent past. With the sharp increase in documentation requirements for agricultural businesses, the potential of FMIS to become widely used comprehensive management tools is high but not yet fully exploited in practice. Consequently, there is a need to improve and adjust these systems. However, there is little research about how to optimize FMIS in order to meet the needs of the users. To answer this question, a deeper understanding of how users differ from other groups regarding their intensity of use and their perceptions of FMIS is needed but has hardly been explored so far. By means of a standardized online questionnaire in the summer of 2019, the skills and attitudes of 280 German farmers regarding the use of FMIS were surveyed. By means of a cluster analysis, different user segments were analyzed to gain deeper insights into which characteristics FMIS users and non-users have. Four clusters were identified including two non-user groups and two user groups with varying potential for FMIS providers and marketers. The results show that the lack of persuasiveness of the existing systems by not (yet) offering solutions to farmers’ problems and needs is a major barrier. One main driver for the use of FMIS is to manage the increasing political and social requirements with the help of a professional documentation of farm activities. On the one hand, the results accompany the ongoing digital transformation in agribusiness and contribute on the other hand to the digitization process on farms by increasing new knowledge about farmers’ requirements regarding FMIS. The results also provide relevant insights for software development companies to adjust their products according to the farmers’ needs.
查看更多>>摘要：? 2022 Elsevier B.V.Fractional vegetation cover (FVC) plays an important role in the study of vegetation growth state, and the key issue is accurately segmenting and extracting green vegetation from the background. However, the shadows generated by natural lights produce extreme illuminance differences in images, which greatly reduces the vegetation extraction accuracy. The polarization information for ground objects is independent of the physical state of the reflectivity of ground objects, and it can be used to eliminate the influence of strong reflections in images to a certain extent, reduce the illuminance differences under extreme sunlight conditions, and help improve the vegetation recognition effect under shadow conditions. To improve the accuracy of vegetation segmentation under shadow conditions, this study introduces polarized reflection information for vegetation and an improved semantic segmentation network, notably a double input residual network based on DeepLabv3plus (DIR_DeepLabv3plus), with fusion strategies based on concatenation and addition is proposed. The network extracts low-level features and high-level features at different spatial scales from both light intensity (red-greenblue (RGB)) images and degree of linear polarization (DoLP) images independently by a deep residual network and atrous spatial pyramid pooling (ASPP) structure, effectively improving the accuracy of vegetation segmentation in shadow situations. The results show that the mean intersection over union (mIoU) values of vegetation without shadows, with light shadows and with shadows are 94.01%, 92.508% and 90.969%, respectively. Compared with the color index method and green fractional vegetation cover extraction from digital images using a shadow-resistant algorithm (SHAR-LABFVC), the proposed method provides a greatly improved extraction accuracy, and it has 0.18%, 1.00% and 1.49% higher mIoU values for vegetation under different shadow conditions than the method without polarization information. This study provides a new approach for vegetation segmentation and improves the accuracy of FVC calculations under shadow conditions.
查看更多>>摘要：? 2022 The AuthorsGrazing livestock contributes to pasture nitrogen (N) through urine and faeces and N losses in pasture-based livestock systems are recognized as an important consideration for sustainable land management. Knowing the frequency and quantity of excreta produced in pasture-based dairy production systems could be useful for informing best management practice. The aim of this experiment was to determine whether data from tail-mounted accelerometers could be used to classify dairy cow excretory events. Ten non-lactating Holstein dairy cows were fitted with a tail-mounted accelerometer set to record data at 1 Hz and were individually observed for 3.3–5.3 h each. The recorded behaviours were urination, defecation, standing and lying (both left and right laterality). G-force acceleration values for X, Y and Z axes were downloaded and windows of varying sizes (3-, 6-, 9-, 12- and 15 s) were used to extract a set of basic features (mean, minimum, maximum and SD) from consecutive sequences of each behaviour. Windows were stepped forward by 1 s for feature extraction and five datasets were developed. Data for all cows were compiled and a random forest algorithm was used for model development. Ten times 10-fold stratified cross-validation (SCV) was used to evaluate data from all window sizes. Sensitivity and precision always exceeded 84% for standing and lying postures in both unbalanced and balanced datasets. Classification performance for excretory events improved significantly (P < 0.01) as window size increased. Due to performance, the 15 s window was selected for further tests with a full feature set. Random forest models were developed using a leave-one-cow-out cross-validation strategy (model developed using n-1 cows and evaluated on the held-out cow). Classification performance for standing and lying remained high (sensitivity & precision ≥ 91 %) but performance for excretory events was poor and highly variable. SCV results for excretory events were clearly optimistic and more data are needed for further testing. However, it may also be necessary to develop and test individual animal models for comparison because there may be considerable variation between animals for excretory events.
查看更多>>摘要：? 2022 Elsevier B.V.As the traditional indicators of freshness measurement of meat products, TVB-N and TBA have the disadvantage of time-consuming, labor-intensive and destructive to the sample. The objective of this study was to investigate the possibility of computer vision techniques to visualize the variation of TVB-N and TBA during the storage of smoked chicken thighs. In this study, freshness indicators (TVB-N and TBA) and images of smoked chicken thighs were obtained simultaneously every 3 days during storage at 4 °C. Then, the RGB color space was converted to HSI and L*a*b* color spaces by color conversion algorithm, and the color parameters (RGB, HSI and L*a*b*) were correlated with TVB-N and TBA, respectively, for establishing multiple regression models. Finally, visualization maps of the spoilage were established by applying the multiple regression model to each pixel in the image. The results showed that the multiple linear regression models of TBA and TVB-N based on the color parameters L*, a*, I, S and R were well correlated (R 2 = 0.993 for TBA and R 2 = 0.970 for TVB-N). Distribution maps of TBA and TVB-N changed color gradually from blue to red during storage, respectively. In conclusion, this study demonstrated that distribution maps can be employed as a rapid, objective, and non-destructive method to predict the TBA and TVB-N values of smoked chicken thighs during storage.
查看更多>>摘要：? 2022 Elsevier B.V.Longan is a famous speciality fruit and cultivated medicinal plant that has important edible and medicinal value; how to improve productivity in harvest is an important issue. At present, longan is mainly planted in hilly areas. For complex site conditions and tall trees, the ground harvesting machineries cannot work normally. In this study, aiming at harvesting longan fruit using unmanned aerial vehicles, a method combining an improved YOLOv5s, improved DeepLabv3+ model and depth image information is proposed, which is used for the three-dimensional (3D) positioning of branch picking points in complex natural environments. First, the improved YOLOv5s model is used to quickly detect longan fruit skewers and the main fruit branches from a complex orchard environment. The correct main fruit branch is obtained according to its relative position relationship and is extracted as the input to the semantic segmentation model. Second, using the improved DeepLabv3+ model, the image extracted in the previous step is semantically segmented to obtain the 2D coordinate information of the main longan fruit branches. Finally, combined with the growth characteristics of a longan fruit string, RGB-D information fusion is carried out on the main fruit branches in 3D space to obtain the central axis and pose information of the main fruit branches, and the 3D coordinates of the picking points are calculated, which provides destination information for a longan harvesting drone. To verify the effectiveness of the proposed method, an experiment for identifying and locating the main fruit branches and picking points was carried out in a longan orchard. The experimental results show that the longan string fruit and main fruit branch detection accuracy is 85.50%, and the main fruit branch semantic segmentation accuracy is 94.52%. The whole algorithm takes 0.58 s in the actual scene and can quickly and accurately locate the picking points. In summary, this paper fully exploits the advantages of the combination of a convolutional neural network and RGB-D image information, further improving the efficiency of longan harvesting drones in accurately positioning picking points in 3D space.
查看更多>>摘要：? 2022 Elsevier B.V.Crop yield is greatly impacted by climate change, and a systemic assessment of its impacts on crop yields is essential. Aiming to investigate the impact of climate change on spring and summer maize yields in main maize growing areas of China, the observed meteorological and maize yield data over 1988–2017 at the 121 sites (including 85 sitess for spring maize and 36 sites for summer maize) in main maize growing areas of China were collected. The first-order difference, Sen's slopes and trend test, multi-collinearity detection, Pearson correlation, stepwise linear and nonlinear regression methods were used, and the best statistical regression models between maize yield and climate variables have been established. Of these, the Sen′s slopes quantify the trend magnitude of the related climate variables and spring/summer maize yields during maize growth period. The Pearson correlation coefficients assess the relationship between pairs of climatic variables and maize yields, while the multi-collinearity analysis determines the mutually independent climatic variables with maize yields. The stepwise multi-variate linear and nonlinear regressions were conducted to obtain the best functions of the one-order-differences of spring (summer)maize yields at the 85 (36) sites. The results indicated that: (1) Generally, the precipitation and temperature during growth seasons was rising, while relative air humidity and sunshine hours was declining. Both the yields of spring and summer maize showed increasing trends. (2) Spring maize yields were more related to relative humidity, sunshine hours and precipitation, while summer maize yields were more related to precipitation and temperature. (3) The multivariate nonlinear functions performed better than the linear relationship. Based on the coefficient of determination, climate change has explained 5.8–87.6% variability of spring maize yield and 6.6–78.5% variability of summer maize yield. (4) The contribution importance rank of climate variables to yields of spring and summer maize was precipitation > relative humidity > sunshine hours > minimum temperature > maximum temperature > average temperature. The wet-cold and wet-warm climate, especially the former, had positive effects on maize yield. In conclusion, climate variables affect spring and summer maize yields and their best relationships were site-specific in China. Our research provides new insights for maize planting management under climate change.
查看更多>>摘要：? 2022 The AuthorsThe adoption of new sensors for crop monitoring is leading to the acquisition of large amounts of data, which usually are not directly usable for agricultural applications. The 3D point cloud maps of fields and parcels, generated from remotely sensed data, are examples of such big data, which require the development of specific algorithms for their processing and interpretation, with the final aim to obtain valuable information about crop status. This manuscript presents an innovative 3D point cloud processing algorithm for vine row detection and localisation within vineyard maps, based on the detection of key points and a density-based clustering approach. Vine row localisation is a crucial phase in the interpretation of the complex and huge 3D point clouds of agricultural environments, which makes it possible to move the focus from a macro level (parcel and plot scale) to a micro level (plants, fruits and branches). The algorithm outputs fully describe the spatial location of each vine row within the whole 3D model of the agricultural environment by a set of key points and an interpolating curve. The algorithm is specifically conceived to be robust and: (i) independent of the adopted airborne sensor used to acquire the in-field data (not requiring a model with colour or spectral information); (ii) able to manage vineyards with any vine row layout or orientation (such as curvilinear) and (iii) not hindered by the occurrence of missing plants. The experimental results, obtained by processing the models of seven case study parcels, proved the algorithm's reliability and accuracy: the automatic vine row detection was found to be 100% in accordance with the manual one; and the obtained localisation indices showed an average error of 12 cm and standard deviation of 10 cm, which is fully compatible with the considered agricultural applications. In addition, the algorithm outputs can be profitably exploited for enhanced path planning of autonomous agricultural machines adopted for in-field operations.
查看更多>>摘要：? 2022 Elsevier B.V.Fruit harvesting is time-consuming and laborious. Robots can harvest fruits with a high degree of automation, greatly reducing labor requirements, but are limited by their efficiency and high costs. In this research, a prototype apple harvesting robot was designed and constructed. The robotic prototype integrated a binocular camera, a 4-degree-of-freedom manipulator, a vacuum-based end-effector, and a mobile vehicle. The robot detected, positioned, gripped, detached, and placed apples. A manipulator controller was designed to realize rapid control execution. Picking experiments were conducted in a spindle apple orchard. Picking was tested using rotation-pull and pull patterns. The test results showed that the rotation-pull pattern was more effective picking apples than the pull pattern. The picking success rate of the rotation-pull pattern was 47.37% in the field orchard and 78% in the simulated orchard environment, with picking cycle time of ～ 4 s. The stem damage rate in the field orchard was 11.11%. The developed picking prototype realized the task of picking apples. Its primary advantage was its competitive picking cycle time, which provides a solid foundation for the future improvement of harvesting efficiency.
NSTL
Elsevier