Recognizing terrain features on terrestrial surface using a deep learning model

Li, Wenwen; Zhou, Bin; Hsu, Chia Yu; Li, Yixing; Ren, Fengbo

doi:10.1145/3149808.3149814

Cited by 31 publications

(15 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Interestingly, a number of studies attempt to map features that are at least partially characterized by geomorphic and terrain characteristics using spectral data only, without using terrain data. For example, Li et al [49] mapped craters from image data using faster R-CNN and obtained a mean average precision (mAP) higher than 0.90. As an example of a study that combined spectral and terrain data, Ghorbanzadeh et al [50] used RapidEye satellite data and measures of plan curvature, topographic slope, and topographic aspect to detect landslides.…”

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

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%

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

“…The combination of these types of erosion contributes to the appearance of various loess landform types (Kale & Gupta, 2001). Vertical erosion can increase the depth of gullies, lateral erosion can expand gully channels and headward erosion can gradually extend gullies and cause the retreat of gully heads (Li et al, 2017). The construction of check dams raises the local erosion base that greatly influences erosion in upstream areas and has great significance with respect to erosion on the gully channel.…”

Section: Gully Erosion and Check Damsmentioning

confidence: 99%

“…Given their powerful abilities to extract information, deep learning methods are used to complete complex segmentation and classification tasks on the basis of the extraction of high-level features (Hamylton et al, 2020;Li & Hsu, 2020). Moreover, these approaches outperform traditional methods when classifying low-level features Li, Zhou, Hsu, Li, & Ren, 2017;Nogueira, Penatti, & Dos Santos, 2017). However, due to the homogeneity of the objects and the backgrounds (Chen et al, 2019), deep learning methods demonstrate low boundary accuracy, especially when extracting natural objects.…”

Section: Introductionmentioning

confidence: 99%

Extracting check dam areas from high‐resolution imagery based on the integration of object‐based image analysis and deep learning

Xiong

et al. 2021

Land Degrad Dev

View full text Add to dashboard Cite

Soil loss is a global environmental problem that can intensively damage surrounding ecosystems. To control soil loss and secure agricultural activities, check dams are constructed for soil conservation. However, due to ineffective management, many check dams are abandoned and are highly prone to be damaged due to rainstorms. Such a phenomenon would cause more serious damage to surrounding environments than that associated with common soil loss. The similar basic signatures of check dam areas and their surroundings can blur the boundaries of these structures in images and negatively affect boundary identification, thereby limiting the effectiveness of traditional check dam area extraction techniques based on the pixel level or visual inspection. To facilitate the extraction of check dams, we propose a method that integrates deep learning and object‐based image analysis. We select the Loess Plateau, on which several effective check dam systems have been constructed in recent years to address intense soil loss, as the study area on which to perform high‐resolution imagery experiments to determine the influences of different sample combinations. The parameters influencing the segmentation algorithm are also examined to determine the best parameter combination for the extraction of check dams. Four test areas comprising 12 check dams across different environments were selected with which to test the accuracy of the proposed method. In addition, we compared the check dam extraction capabilities of our proposed method with those of the random forest and deep learning approaches. The results show that the proposed method can achieve a classification accuracy and kappa coefficient that signify good performance in detecting the boundaries and areas of check dams. The proposed method generally outperforms the random forest and deep learning techniques. The extraction results can support the efficient soil management and guide future studies on gully erosion.

show abstract

“…Effective large aerial imagery datasets processing is of fundamental importance for many applications, including maps creation, land use mapping, geological processes, navigation, and place-based studies. W. Li et al [30], [31] utilized CNN to detect terrain features such as craters, lakes, volcanos, and sand dunes. The model was trained with remote sensing imagery and image augmentation and ensemble learning techniques were used for training.…”

Section: Remotely Sensed Imagery Processingmentioning

confidence: 99%

Data Acquisition and Processing for GeoAI Models to Support Sustainable Agricultural Practices

Ojo¹,

Curry²,

Porwol³

2020

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

There are growing opportunities to leverage new technologies and data sources to address global problems related to sustainability, climate change, and biodiversity loss. The emerging discipline of GeoAI resulting from the convergence of AI and Geospatial science (Geo-AI) is enabling the possibility to harness the increasingly available open Earth Observation data collected from different constellations of satellites and sensors with high spatial, spectral and temporal resolutions. However, transforming these raw data into high-quality datasets that could be used for training AI and specifically deep learning models are technically challenging. This paper describes the process and results of synthesizing labelled-datasets that could be used for training AI (specifically Convolutional Neural Networks) models for determining agricultural land use pattern to support decisions for sustainable farming. In our opinion, this work is a significant step forward in addressing the paucity of usable datasets for developing scalable GeoAI models for sustainable agriculture.

show abstract

Recognizing terrain features on terrestrial surface using a deep learning model

Cited by 31 publications

References 13 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

Extracting check dam areas from high‐resolution imagery based on the integration of object‐based image analysis and deep learning

Data Acquisition and Processing for GeoAI Models to Support Sustainable Agricultural Practices

Contact Info

Product

Resources

About