Comparing U-Net convolutional networks with fully convolutional networks in the performances of pomegranate tree canopy segmentation

Zhao, Tiebiao; Yang, Yonghuan; Niu, Haoyu; Wang, Dong; Chen, YangQuan

doi:10.1117/12.2325570

Cited by 70 publications

(50 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, the introduction of fully convolutional neural networks (FCNs) alleviated this limitation by combining convolution and deconvolution layers with up-sampling, which allows for the final feature map to be produced at the original image resolution with a prediction at each cell location [27,33], similar to traditional remote sensing classification products. Example FCN architectures include SegNet [34] and UNet [35][36][37].…”

Section: Deep Learningmentioning

confidence: 99%

“…Zhang et al [51] assessed the method for mapping artic ice-wedge polygons from high spatial resolution aerial imagery and documented that 95% of individual ice-wedge polygons were correctly delineated and classified, with an overall accuracy of 79%. Zhao et al [37] found that Mask R-CNN outperformed UNet for pomegranate tree canopy segmentation. Stewart et al [52] used the method to detect lesions on maize plants from northern leaf blight using unmanned aerial vehicle (UAV) data.…”

Section: Deep Learningmentioning

confidence: 99%

See 1 more Smart Citation

Mapping the Topographic Features of Mining-Related Valley Fills Using Mask R-CNN Deep Learning and Digital Elevation Data

2020

View full text Add to dashboard Cite

Modern elevation-determining remote sensing technologies such as light-detection and ranging (LiDAR) produce a wealth of topographic information that is increasingly being used in a wide range of disciplines, including archaeology and geomorphology. However, automated methods for mapping topographic features have remained a significant challenge. Deep learning (DL) mask regional-convolutional neural networks (Mask R-CNN), which provides context-based instance mapping, offers the potential to overcome many of the difficulties of previous approaches to topographic mapping. We therefore explore the application of Mask R-CNN to extract valley fill faces (VFFs), which are a product of mountaintop removal (MTR) coal mining in the Appalachian region of the eastern United States. LiDAR-derived slopeshades are provided as the only predictor variable in the model. Model generalization is evaluated by mapping multiple study sites outside the training data region. A range of assessment methods, including precision, recall, and F1 score, all based on VFF counts, as well as area- and a fuzzy area-based user’s and producer’s accuracy, indicate that the model was successful in mapping VFFs in new geographic regions, using elevation data derived from different LiDAR sensors. Precision, recall, and F1-score values were above 0.85 using VFF counts while user’s and producer’s accuracy were above 0.75 and 0.85 when using the area- and fuzzy area-based methods, respectively, when averaged across all study areas characterized with LiDAR data. Due to the limited availability of LiDAR data until relatively recently, we also assessed how well the model generalizes to terrain data created using photogrammetric methods that characterize past terrain conditions. Unfortunately, the model was not sufficiently general to allow successful mapping of VFFs using photogrammetrically-derived slopeshades, as all assessment metrics were lower than 0.60; however, this may partially be attributed to the quality of the photogrammetric data. The overall results suggest that the combination of Mask R-CNN and LiDAR has great potential for mapping anthropogenic and natural landscape features. To realize this vision, however, research on the mapping of other topographic features is needed, as well as the development of large topographic training datasets including a variety of features for calibrating and testing new methods.

show abstract

Section: Deep Learningmentioning

confidence: 99%

Section: Deep Learningmentioning

confidence: 99%

Mapping the Topographic Features of Mining-Related Valley Fills Using Mask R-CNN Deep Learning and Digital Elevation Data

2020

View full text Add to dashboard Cite

show abstract

“…Semantic segmentation, different from image level classification (generating only one label for the whole input image), is to classify input image into a number of class labels for each pixel. This technique is especially preferred in applications such as remote sensing [13] and biomedical image analysis [14]. Traditional ways for semantic segmentation include point, line and edge detection methods, thresholding, region-based, pixel-based clustering and morphological approaches.…”

Section: Introductionmentioning

confidence: 99%

“…Very recently, the state-of-the-art U-Net is applied in [10] for leaf level disease segmentation of cucumber leaf with promising performance. U-Net and mask R-CNN [16] are compared [13] for tree canopy segmentation by using UAV RGB image at 30m. To summarize, the following observations are presented to motivate the research in this study: (i) RGB image only possesses three visible bands (Blue, Green and Red), and its image quality is easily susceptible to environmental variations [9] in comparison to multispectral image with an accurate calibration panel; (ii) Disease monitoring based on purely spectral information [7], may lead to a high proportion of false positives due to the spatial inhomogeneity; (iii) Pixel-level CNN is effective in extracting spectral-spatial features [6], [15], however, patch size is empirically determined and it also involves a high computation load; (iv) Semantic segmentation based on FCN (e.g.…”

Section: Introductionmentioning

confidence: 99%

Aerial Visual Perception in Smart Farming: Field Study of Wheat Yellow Rust Monitoring

et al. 2021

IEEE Trans. Ind. Inf.

126

View full text Add to dashboard Cite

Agriculture is facing severe challenges from crop stresses, threatening its sustainable development and food security. This work exploits aerial visual perception for yellow rust disease monitoring, which seamlessly integrates state-of-the-art techniques and algorithms including UAV sensing, multispectral imaging, vegetation segmentation and deep learning U-Net. A field experiment is designed by infecting winter wheat with yellow rust inoculum, on top of which multispectral aerial images are captured by DJI Matrice 100 equipped with RedEdge camera. After image calibration and stitching, multispectral orthomosaic is labelled for system evaluation by inspecting highresolution RGB images taken by Parrot Anafi Drone. The merits of the developed framework drawing spectral-spatial information concurrently are demonstrated by showing improved performance over purely spectral based classifier by the classical random forest algorithm. Moreover, various network input band combinations are tested including three RGB bands and five selected spectral vegetation indices by Sequential Forward Selection strategy of Wrapper algorithm.

show abstract

“…Carlos et al [26] applied four segmentation methods (k-means, artificial neural network (ANN), random forest (RForest), and spectral indices (SI)) to grape tree canopies and found that the SI+ANN and RForest methods were superior, with an accuracy of approximately 0.98 in high-resolution UAV images of trees under different shade and soil conditions, useful for the exquisite management of commercial vineyards. Zhao et al [27] segmented regions of pomegranate trees using the U-net and a region-CNN with a high-resolution visible imaging and multi-spectral imaging UAV system and calculated the water stress parameter and nutritional status of the trees in multi-spectral data. Ultimately, they found that the region-CNN provided better segmentation results.…”

Section: Introductionmentioning

confidence: 99%

Citrus Tree Segmentation from UAV Images Based on Monocular Machine Vision in a Natural Orchard Environment

Chen

Hou

Tang

et al. 2019

Sensors

View full text Add to dashboard Cite

The segmentation of citrus trees in a natural orchard environment is a key technology for achieving the fully autonomous operation of agricultural unmanned aerial vehicles (UAVs). Therefore, a tree segmentation method based on monocular machine vision technology and a support vector machine (SVM) algorithm are proposed in this paper to segment citrus trees precisely under different brightness and weed coverage conditions. To reduce the sensitivity to environmental brightness, a selective illumination histogram equalization method was developed to compensate for the illumination, thereby improving the brightness contrast for the foreground without changing its hue and saturation. To accurately differentiate fruit trees from different weed coverage backgrounds, a chromatic aberration segmentation algorithm and the Otsu threshold method were combined to extract potential fruit tree regions. Then, 14 color features, five statistical texture features, and local binary pattern features of those regions were calculated to establish an SVM segmentation model. The proposed method was verified on a dataset with different brightness and weed coverage conditions, and the results show that the citrus tree segmentation accuracy reached 85.27% ± 9.43%; thus, the proposed method achieved better performance than two similar methods.

show abstract

Comparing U-Net convolutional networks with fully convolutional networks in the performances of pomegranate tree canopy segmentation

Cited by 70 publications

References 14 publications

Mapping the Topographic Features of Mining-Related Valley Fills Using Mask R-CNN Deep Learning and Digital Elevation Data

Mapping the Topographic Features of Mining-Related Valley Fills Using Mask R-CNN Deep Learning and Digital Elevation Data

Aerial Visual Perception in Smart Farming: Field Study of Wheat Yellow Rust Monitoring

Citrus Tree Segmentation from UAV Images Based on Monocular Machine Vision in a Natural Orchard Environment

Contact Info

Product

Resources

About