气候变化领域集成服务门户(CCPortal): Adopting deep learning methods for airborne RGB fluvial scene classification

CCPortal

DOI	10.1016/j.rse.2020.112107
	Adopting deep learning methods for airborne RGB fluvial scene classification
	Carbonneau P.E.; Dugdale S.J.; Breckon T.P.; Dietrich J.T.; Fonstad M.A.; Miyamoto H.; Woodget A.S.
发表日期	2020
ISSN	00344257
卷号	251
英文摘要	Rivers are among the world's most threatened ecosystems. Enabled by the rapid development of drone technology, hyperspatial resolution (<10 cm) images of fluvial environments are now a common data source used to better understand these sensitive habitats. However, the task of image classification remains challenging for this type of imagery and the application of traditional classification algorithms such as maximum likelihood, still in common use among the river remote sensing community, yields unsatisfactory results. We explore the possibility that a classifier of river imagery based on deep learning methods can provide a significant improvement in our ability to classify fluvial scenes. We assemble a dataset composed of RGB images from 11 rivers in Canada, Italy, Japan, the United Kingdom, and Costa Rica. The images were labelled into 5 land-cover classes: water, dry exposed sediment, green vegetation, senescent vegetation and roads. In total, >5 billion pixels were labelled and partitioned for the tasks of training (1 billion pixels) and validation (4 billion pixels). We develop a novel supervised learning workflow based on the NASNet convolutional neural network (CNN) called ‘CNN-Supervised Classification’ (CSC). First, we compare the classification performance of maximum likelihood, a multilayer perceptron, a random forest, and CSC. Results show median F1 scores (a commonly used quality metric in machine learning) of 71%, 78%, 72% and 95%, respectively. Second, we train our classifier using data for 5 of 11 rivers. We then predict the validation data for all 11 rivers. For the 5 rivers that were used in model training, median F1 scores reach 98%. For the 6 rivers not used in model training, median F1 scores are 90%. We reach two conclusions. First, in the traditional workflow where images are classified one at a time, CSC delivers an unprecedented mix of labour savings and classification F1 scores above 95%. Second, deep learning can predict land-cover classifications (F1 = 90%) for rivers not used in training. This demonstrates the potential to train a generalised open-source deep learning model for airborne river surveys suitable for most rivers ‘out of the box’. Research efforts should now focus on further development of a new generation of deep learning classification tools that will encode human image interpretation abilities and allow for fully automated, potentially real-time, interpretation of riverine landscape images. © 2020 Elsevier Inc.
英文关键词	Airborne imagery; Deep learning; Fluvial remote sensing; Land cover classification
语种	英语
scopus关键词	Convolutional neural networks; Decision trees; Image classification; Image enhancement; Learning systems; Maximum likelihood; Multilayer neural networks; Pixels; Remote sensing; Rivers; Supervised learning; Surveys; Vegetation; Classification algorithm; Classification performance; Classification tool; Hyperspatial resolutions; Image interpretation; Land cover classification; Scene classification; Supervised classification; Deep learning; airborne sensing; data set; image classification; machine learning; model validation; remote sensing; supervised learning; Canada; Costa Rica; Italy; Japan; United Kingdom
来源期刊	Remote Sensing of Environment
文献类型	期刊论文
条目标识符	http://gcip.llas.ac.cn/handle/2XKMVOVA/179105
作者单位	Department of Geography, Durham University, Mountjoy Site, Durham, DH1 3LE, United Kingdom; Department of Computer Sciences, Durham University, Mountjoy Site, Durham, DH1 3LE, United Kingdom; School of Geography, University of Nottingham, University Park, Nottingham, NG7 2RD, United Kingdom; Department of Geography, University of Northern Iowa, 1227 West 27th Street Cedar FallsIA 50614, United States; Department of Geography, University of Oregon, 1251 University of Oregon, Eugene, OR 97403-1251, United States; Department of Civil Engineering, Shibaura Institute of Technology, 3-7-5 Toyosu, Koto-ku, Tokyo, 135-8548, Japan; Department of Geography and Environment, Loughborough University, Epinal Way, Loughborough, Leicestershire, LE11 3TU, United Kingdom
推荐引用方式 GB/T 7714	Carbonneau P.E.,Dugdale S.J.,Breckon T.P.,et al. Adopting deep learning methods for airborne RGB fluvial scene classification[J],2020,251.
APA	Carbonneau P.E..,Dugdale S.J..,Breckon T.P..,Dietrich J.T..,Fonstad M.A..,...&Woodget A.S..(2020).Adopting deep learning methods for airborne RGB fluvial scene classification.Remote Sensing of Environment,251.
MLA	Carbonneau P.E.,et al."Adopting deep learning methods for airborne RGB fluvial scene classification".Remote Sensing of Environment 251(2020).