127 research outputs found
Narrative structure of Bidayuh animal tales
No full textChapter 2, by Florence Gilliam Kayad and Palisya Siew-Ching
Ting, unravels the uniqueness of Bidayuh folk tales while at the same time affirming the universality of the narrative structure of the tales. Interestingly, tales with animal-human relationship such as “The Tree of Siburan” have more characters and a more complex plot than tales with
animal-animal relationship such as “The mouse deer and the snail” and tales with human-animal transformation such as “Sibago, the cockerel”. The authors gauged how familiar some younger Bidayuhs were with the folktales using the reader response approach, and found out that they often only knew the gist of the tales. The authors raised a red flag on the
imminent loss of a crucial part of the Bidayuh cultural heritage
Traditional Marriage Customs in Rajasthan, India: A Study of Kayad Village of Ajmer District
No full textAbstract The present research study was conducted in Kayad villag
Grape yield spatial variability assessment using YOLOv4 object detection algorithm
No full textSingle shoot detection algorithms represent a promising tool for real-time application of deep learning models. YOLO (You Only Look Once) is a single shoot object detection algorithm that combines fast classification and good accuracy. Over the last few years, several versions of YOLO have been developed, improving its performance. In this study, the last version of YOLO (version 4) was evaluated in its full-size model and in the tiny version, in order to assess grape yield spatial variability. The tiny and full models were previously trained and tested on almost 3000 images collected during several growing stages in different vineyards and varieties. YOLO models were used to classify 24 georeferenced RGB images acquired before the harvesting on an 8-hectare experimental vineyard, where Vitis vinifera cv. Glera vines were trained to Sylvoz and characterized by spatially structured variability. The models were used to detect the number of bunches, based on different resolution images (from 320 up to 1280 pixels) and different confidence thresholds (from 0.25 up to 0.35). The detected number of bunches was then compared with the actual one as well as with the relative final weight harvested from the vines used as target for the collected images.
According to preliminary results, the number of bunches detected in high-resolution images exhibited a higher correlation with the number of bunches visible in the images rather than with the final weight. On the other hand, the number of bunches detected in low-resolution images gave evidence of a higher correlation with the total weight of grapes harvested from the target vines. Although high-resolution images allowed the model to detect almost all bunches not covered by leaves, in low-resolution images YOLO models were weakly affected by small bunches, which were rarely detected, thus increasing the correlation with the vines yield. The best linear regression model for vines yield was obtained with 416 pixels images, which showed a coefficient of determination (R2) of 0.59, indicating YOLO as a suitable tool for detecting yield spatial variability. The models used in this work represent non-destructive methodologies for grape yield spatial variability assessment, and they may be easily implemented as on-the-go tools
Automatic Bunch Detection in White Grape Varieties Using YOLOv3, YOLOv4, and YOLOv5 Deep Learning Algorithms
No full textOver the last few years, several Convolutional Neural Networks for object detection have been proposed, characterised by different accuracy and speed. In viticulture, yield estimation and prediction is used for efficient crop management, taking advantage of precision viticulture techniques. Convolutional Neural Networks for object detection represent an alternative methodology for grape yield estimation, which usually relies on manual harvesting of sample plants. In this paper, six versions of the You Only Look Once (YOLO) object detection algorithm (YOLOv3, YOLOv3-tiny, YOLOv4, YOLOv4-tiny, YOLOv5x, and YOLOv5s) were evaluated for real-time bunch detection and counting in grapes. White grape varieties were chosen for this study, as the identification of white berries on a leaf background is trickier than red berries. YOLO models were trained using a heterogeneous dataset populated by images retrieved from open datasets and acquired on the field in several illumination conditions, background, and growth stages. Results have shown that YOLOv5x and YOLOv4 achieved an F1-score of 0.76 and 0.77, respectively, with a detection speed of 31 and 32 FPS. Differently, YOLO5s and YOLOv4-tiny achieved an F1-score of 0.76 and 0.69, respectively, with a detection speed of 61 and 196 FPS. The final YOLOv5x model for bunch number, obtained considering bunch occlusion, was able to estimate the number of bunches per plant with an average error of 13.3% per vine. The best combination of accuracy and speed was achieved by YOLOv4-tiny, which should be considered for real-time grape yield estimation, while YOLOv3 was affected by a False Positive–False Negative compensation, which decreased the RMSE
wGrapeUNIPD-DL: An open dataset for white grape bunch detection
No full textNational and international Vitis variety catalogues can be used as image datasets for computer vision in viticulture. These databases archive ampelographic features and phenology of several grape varieties and plant structures images (e.g. leaf, bunch, shoots). Although these archives represent a potential database for computer vision in viticulture, plant structure images are acquired singularly and mostly not directly in the vineyard. Localization computer vision models would take advantage of multiple objects in the same image, allowing more efficient training. The present images and labels dataset was designed to overcome such limitations and provide suitable images for multiple cluster identification in white grape varieties. A group of 373 images were acquired from later view in vertical shoot position vineyards in six different Italian locations at different phenological stages. Images were then labelled in YOLO labelling format. The dataset was made available both in terms of images and labels. The real number of bunches counted in the field, and the number of bunches visible in the image (not covered by other vine structures) was recorded for a group of images in this dataset
Comparing maize leaf area index retrieval from aerial hyperspectral images through radiative transfer model inversion and machine learning techniques
No full textThis study compares maize leaf area index (LAI) retrieval methods based on radiative transfer models and machine learning techniques. Ground LAI was measured from the study field at different growth stages where aerial hyperspectral images were acquired at the same stages. The PROSAIL-based model was built using a range of maize leaf and canopy parameters with a total of >21k simulations covering different maize spectral reflectance scenarios. Moreover, random forest and support vector machines were applied to the spectral and corresponding ground LAI measurements divided into 50% for model training and 50% for validation. Results showed that the PROSAIL-based model provided the highest R2 value between ground and estimated LAI followed by the RF and SVM where R2 values were 0.65, 0.59 and 0.35 respectively
Impact of soil firmness and tillage depth on irrigated maize silage performance
No full textAbstract. A field experiment was conducted to investigate the response of maize silage ( L.) to tillage depth under different soil firmness levels. The study was carried out on a 16 ha center-pivot irrigated field in a commercial farm located in the eastern region of Saudi Arabia. A soil firmness map was generated and used as a management map. This map was divided into three soil firmness zones based on soil cone index (low: 617 to 1270 kPa for a 0 to 15 cm depth in undisturbed soil, medium: 1271 to 1652 kPa and high: 1653 to 2306 kPa). Three tillage depth treatments (10, 20, and 25 cm) were imposed on each of the three soil firmness zones, using a tandem disc harrow. Maize growth parameters [plant population, plant height, and Normalized Difference Vegetation Index (NDVI)] and maize silage yield were used to evaluate the response of the maize crop to tillage depth. The results revealed that soil firmness and tillage depth at both early (25 days after sowing) and late (60 days after sowing) growth stages did not introduce significant effects on maize plant population. However, the plant height measured at 60 days after sowing showed a significant response to soil firmness. The lowest mean value of plant height (114.4 cm) was recorded at the high soil firmness level, while the greatest mean value (136.3 cm) was recorded under low soil firmness level. Also, significant differences in maize silage yield were recorded under different soil firmness levels and tillage depths. For maize silage production, a tillage depth of 10 cm was observed to be optimum for areas of low and medium soil firmness. For areas of high soil firmness, the optimum tillage depth was 20 cm. Keywords: Disc harrow, Kriging, Maize, Maps, Penetrometer, Tillage depths.</jats:p
Validation of a commercial optoelectronics device for grape quality analysis
No full textIn modern viticulture, grape quality monitoring and management during ripening plays a key role in the winemaking process. Conventional monitoring techniques are based on in-field refractometer measurement and eventually determination of acidity by titration, which are time-consuming tasks. Despite proximal and remote sensing may be used to estimate grape maturity in a relatively short time, the determination of the number of specific chemical components that change during the maturation process, such as malic acid and tartaric acid, may be not accurate. Smart chemometric techniques represent promising technologies for grape quality monitoring, allowing accurate grape chemical components determination in a short time. In the present study, a novel optoelectronic device for colorimetric and enzymatic analysis is presented: it relies on implementation of a LED light source with a wavelength emission ranging between 340 nm and 760 nm, a cuvette housing containing berry juice, and a photodiode. The device was applied for the determination of grape components involved in the maturation process: ten sampling plots of 20 plants each were selected in a 9-hectare experimental vineyard (Vitis vinifera cv. Glera trained as Sylvoz). The maturation development was monitored by collecting 250 berries from each sampling plot at 3 different times, from veraison until harvesting. Berries juice was analyzed using the optoelectronic device along with the conventional methodology. Absorbance measurements were used in order to determine the grape concentration of dissolved sugar, malic acid, and tartaric acid. In addition, the device was implemented to determine the pH and titratable acidity of the grape sample. Preliminary results gave evidence of an RMSE of 0.119 and 1.027 and an R2 of 0.94 and 0.98 for pH and titrable acidity respectively, suggesting an effective use of the optoelectronic device for grape quality monitoring. The final paper will include dissolved sugar content, malic acid, and tartaric acid validation using HPLC methodology
Latest advances in sensor applications in agriculture
No full textSensor applications are impacting the everyday objects that enhance human life quality. In this special issue, the main objective was to address recent advances of sensor applications in agriculture covering a wide range of topics in this field. A total of 14 articles were published in this special issue where nine of them were research articles, two review articles and two technical notes. The main topics were soil and plant sensing, farm management and post-harvest application. Soil-sensing topics include monitoring soil moisture content, drain pipes and topsoil movement during the harrowing process while plant-sensing topics include evaluating spray drift in vineyards, thermography applications for winter wheat and tree health assessment and remote-sensing applications as well. Furthermore, farm management contributions include food systems digitalization and using archived data from plowing operations, and one article in post-harvest application in sunflower seeds
- …
