A UAV Open Dataset of Rice Paddies for Deep Learning Practice

Yang, Ming‐Der; Tseng, Hsin-Hung; Hsu, Yu‐Chun; Yang, Chin-Ying; Lai, Ming-Hsin; Wu, Dong-Hong

doi:10.3390/rs13071358

Cited by 22 publications

(5 citation statements)

References 33 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The majority of computer vision approaches depend on extensive datasets for the purpose of training, testing, and assessing different solutions to issues [9]. They offer the resources to educate and assess novel algorithms, facilitating a direct comparison of the outcomes [10], [11]. Ultimately, they enable researchers to address novel and increasingly complex research problems [12].…”

Section: Related Workmentioning

confidence: 99%

A dataset for computer-vision-based fig fruit detection in the wild with benchmarking you only look once model detector

Che Ani,

Hamdani Mohammad Farid,

Firdhaus Kamaruzaman

et al. 2024

Bulletin EEI

View full text Add to dashboard Cite

The image datasets that are most widely used for training deep learning models are specifically developed for applications. This study introduces a novel dataset aimed at augmenting the existing data for the identification of figs in their natural habitats, specifically in the wilderness. In the present study, researchers have generated numerous image datasets specifically for object detection focus on applications in agriculture. Regrettably, it is exceedingly difficult for us to obtain a specialized dataset specifically designed for detecting figs. To tackle this issue, a grand total of 462 photographs of fig fruits were gathered. The augmentation technique was utilized to substantially increase the size of the dataset. Ultimately, we conduct an examination of the dataset by doing a baseline performance study for bounding-box detection using established object detection methods, specifically you only look once (YOLO) version 3 and YOLOv4. The performance obtained on the test photos of our dataset is satisfactory. For farmers, the capacity to identify and oversee fig fruits in their natural or developed environments can be highly advantageous. The detecting device offers instantaneous data regarding the quantity of mature figs, facilitating decision-making procedures.

show abstract

Section: Related Workmentioning

confidence: 99%

A dataset for computer-vision-based fig fruit detection in the wild with benchmarking you only look once model detector

Che Ani,

Hamdani Mohammad Farid,

Firdhaus Kamaruzaman

et al. 2024

Bulletin EEI

View full text Add to dashboard Cite

show abstract

“…The deep learning semantic segmentation model is pixel-based image classification, which aims to perform pixel-by-pixel recognition on an image and annotate each pixel with a class label [45]. The classification system used in this study was grape, peach, apple, cherry, corn, other trees, and other features (any ground object except for the other six features).…”

Section: Sample Set Construction and Sample Labelingmentioning

confidence: 99%

“…RGB labels were assigned to each class, the boundaries of each crop were sketched and colored using Photoshop, the label RGB values are shown in Table 2, and the label map that was made is shown in Figure 6. The deep learning semantic segmentation model is pixel-based image classification, which aims to perform pixel-by-pixel recognition on an image and annotate each pixel with a class label [45]. The classification system used in this study was grape, peach, apple, cherry, corn, other trees, and other features (any ground object except for the other six features).…”

Section: Sample Set Construction and Sample Labelingmentioning

confidence: 99%

Economic Fruit Forest Classification Based on Improved U-Net Model in UAV Multispectral Imagery

Jia

Yang

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Economic fruit forest is an important part of Chinese agriculture with high economic value and ecological benefits. Using UAV multi-spectral images to research the classification of economic fruit forests based on deep learning is of great significance for accurately understanding the distribution and scale of fruit forests and the status quo of national economic fruit forest resources. Based on the multi-spectral remote sensing images of UAV, this paper constructed semantic segmentation data of economic fruit forests, conducted a comparative study on the classification and identification of economic fruit forests of FCN, SegNet, and U-Net classic semantic segmentation models, and proposed an improved ISDU-Net model. The recognition accuracies of the ISDU-Net were 87.73%, 70.68%, 78.69%, and 0.84, respectively, in terms of pixel accuracy, average intersection ratio, frequency weight intersection ratio, and Kappa coefficient, which were 3.19%, 8.90%, and 4.51% higher than the original U-Net model. The results showed that the improved ISDU-Net could effectively improve the learning ability of the model, perform better in the prediction of short sample categories, obtain a higher classification accuracy of fruit forest crops, and provide a new idea for the research on accurate fruit forest identification.

show abstract

“…In recent decades, advances in UAS have made this technology a promising tool that involves relatively smaller investments (Manfreda et al, 2018;Yang et al, 2021). Also, continued improvements of aerial vehicles and of the imaging and sensing equipment mounted on these platforms classify UAS as alternative environmental monitoring platforms.…”

Section: Remote Sensing Products and Precision Agriculturementioning

confidence: 99%

Remote Sensing Monitoring of Rice Fields: Towards Assessing Water Saving Irrigation Management Practices

Lima

Gerardo

Lima

2021

Front. Remote Sens.

View full text Add to dashboard Cite

Rice cultivation is one of the largest users of the world’s freshwater resources. The contribution of remote sensing observations for identifying the conditions under which rice is cultivated, particularly throughout the growing season, can be instrumental for water, and crop management. Data from different remote sensing platforms are being used in agriculture, namely to detecting anomalies in crops. This is attempted by calculating vegetation indices (VI) that are based on different vegetation reflectance bands, especially those that rely on the Red, Green, and near-infrared bands, such as the Normalised Difference Vegetation Index (NDVI) or the Green Normalised Difference Vegetation Index (GNDVI). However, particular features of different crops and growing conditions justify that some indices are more adequate than others on a case-to-case basis, according to the different vegetation’s spectral signatures. In recent years, a vegetation index related to the Red Edge reflectance band, the Normalised Difference Red Edge (NDRE) has shown potential to be used as a tool to support agricultural management practices; this edge band, by taking a transition position, is very sensitive to changes in vegetation properties. This work, focusing on the rice crop and the application of different irrigation practices, explores the capability of several VIs calculated from different reflectance bands to detect variability, at the plot scale, in rice cultivation in the Lower Mondego region (Portugal). The remote sensing data were obtained from satellite Sentinel-2A imagery and using a multispectral camera mounted on an Unmanned Aerial System (UAS). By comparing several vegetation indices, we found that NDRE is particularly useful for identifying non-homogeneities in irrigation and crop growth in rice fields. Since few satellite sensors are sensible in the Red Edge band and none has the spatial resolution offered by UAS, this study explores the potential of UAS to be used as a useful support information tool in rice farming and precision agriculture, regarding irrigation, and agronomic management.

show abstract

A UAV Open Dataset of Rice Paddies for Deep Learning Practice

Cited by 22 publications

References 33 publications

A dataset for computer-vision-based fig fruit detection in the wild with benchmarking you only look once model detector

A dataset for computer-vision-based fig fruit detection in the wild with benchmarking you only look once model detector

Economic Fruit Forest Classification Based on Improved U-Net Model in UAV Multispectral Imagery

Remote Sensing Monitoring of Rice Fields: Towards Assessing Water Saving Irrigation Management Practices

Contact Info

Product

Resources

About