Automatic land cover classification of geo-tagged field photos by deep learning. (May 2017)

• Record Type: Journal Article
• Authors: Xu, Guang; Zhu, Xuan; Fu, Dongjie; Dong, Jinwei; Xiao, Xiangming
• Is Part Of: Environmental modelling & software. Volume 91(2017)
• Page Start: 127
• Page End: 134
• Publication Date: 2017-05
• Subjects: Deep learning -- Convolutional neural network -- Transfer learning -- Multinomial logistic regression -- Land cover -- Crowdsourced photos; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2017.02.004
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: With more and more crowdsourcing geo-tagged field photos available online, they are becoming a potentially valuable source of information for environmental studies. However, the labelling and recognition of these photos are time-consuming. To utilise such information, a land cover type recognition model for field photos was proposed based on the deep learning technique. This model combines a pre-trained convolutional neural network (CNN) as the image feature extractor and the multinomial logistic regression model as the feature classifier. The pre-trained CNN model Inception-v3 was used in this study. The labelled field photos from the Global Geo-Referenced Field Photo Library (http://eomf.ou.edu/photos ) were chosen for model training and validation. The results indicated that our recognition model achieved an acceptable accuracy (48.40% for top-1 prediction and 76.24% for top-3 prediction) of land cover classification. With accurate self-assessment of confidence, the model can be applied to classify numerous online geo-tagged field photos for environmental information extraction. Highlights: A field photo recognition model based on deep learning technique was proposed. A weighted multinomial logistic regression model was used to deal with uneven sample sizes. The results showed land cover classification of field photos with an acceptable accuracy and high prediction confidence.