A label-noise robust active learning sample collection method for multi-temporal urban land-cover classification and change analysis

Abstract Timely and reliable land-use/land-cover (LULC) change dynamic monitoring is the basis of urban understanding and planning. However, either the training sample shortage or the error accumulation in the multi-temporal processing inevitably restricts the monitoring performance. In this paper, to overcome these problems, we present a label-noise robust active learning method, which automatically collects reliable and informative samples from the images and builds a unified classification system with these augmented samples. In more detail, a Bayesian sample collection process that fuses the unsupervised transition information and the multi-temporal land-cover information is designed to provide candidate samples with “from-to” labels. A reliability-based multi-classifier active learning method is then proposed to adaptively allocate the more reliable samples to the classes that are difficulty to classify. Finally, a fusion of the multiple multi-date classifications trained by the selected samples is implemented to identify the change type of interest. The dynamic monitoring results for Shanghai, Shenzhen, and Shiyan in China, two megacities with rapid and obvious urbanization and a small city with relatively slow urbanization, indicate that the proposed method achieves a significantly higher accuracy than the current state-of-the art methods. The sample accuracy verified by the high spatial resolution reference maps endorses the applicability of the sample collection, while the reliability-based active learning further ensures the robustness of the proposed method in the label-noise situation. The presented method was tested in two difficult situations (a small training sample case and a training sample set without joint labeling), so that the robustness and accuracy of the approach can be expected to be of a similar or better quality in cases with more training samples. Given its effectiveness and robustness, the proposed method could be widely applied in LULC change dynamic monitoring.

[1]  Weiqi Zhou,et al.  A new approach for land cover classification and change analysis: Integrating backdating and an object-based method , 2016 .

[2]  Dengsheng Lu,et al.  Detection of impervious surface change with multitemporal Landsat images in an urban-rural frontier. , 2011, ISPRS journal of photogrammetry and remote sensing : official publication of the International Society for Photogrammetry and Remote Sensing.

[3]  Bo Du,et al.  A post-classification change detection method based on iterative slow feature analysis and Bayesian soft fusion , 2017, Remote Sensing of Environment.

[4]  Zhe Zhu,et al.  Mapping forest change using stacked generalization: An ensemble approach , 2018 .

[5]  Joan Fisher Box,et al.  Guinness, Gosset, Fisher, and Small Samples , 1987 .

[6]  Chengquan Huang,et al.  Use of a dark object concept and support vector machines to automate forest cover change analysis , 2008 .

[7]  Pramod K. Varshney,et al.  An image change detection algorithm based on Markov random field models , 2002, IEEE Trans. Geosci. Remote. Sens..

[8]  D. Stow,et al.  THE EFFECT OF TRAINING STRATEGIES ON SUPERVISED CLASSIFICATION AT DIFFERENT SPATIAL RESOLUTIONS , 2002 .

[9]  Yansheng Li,et al.  The edge-preservation multi-classifier relearning framework for the classification of high-resolution remotely sensed imagery , 2018 .

[10]  Joanne C. White,et al.  Optical remotely sensed time series data for land cover classification: A review , 2016 .

[11]  Hang Zhou,et al.  Deep learning based multi-temporal crop classification , 2019, Remote Sensing of Environment.

[12]  Bin Chen,et al.  Nearest-neighbor diffusion-based pan-sharpening algorithm for spectral images , 2014, Optical Engineering.

[13]  P. Gong,et al.  Tracking annual cropland changes from 1984 to 2016 using time-series Landsat images with a change-detection and post-classification approach: Experiments from three sites in Africa , 2018, Remote Sensing of Environment.

[14]  Lorenzo Bruzzone,et al.  Batch-Mode Active-Learning Methods for the Interactive Classification of Remote Sensing Images , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[15]  Hassiba Nemmour,et al.  Multiple support vector machines for land cover change detection: An application for mapping urban extensions , 2006 .

[16]  J. Fry,et al.  Updating the 2001 National Land Cover Database land cover classification to 2006 by using Landsat imagery change detection methods , 2009 .

[17]  Giles M. Foody,et al.  Good practices for estimating area and assessing accuracy of land change , 2014 .

[18]  John R. Jensen,et al.  A change detection model based on neighborhood correlation image analysis and decision tree classification , 2005 .

[19]  Allan Aasbjerg Nielsen,et al.  The Regularized Iteratively Reweighted MAD Method for Change Detection in Multi- and Hyperspectral Data , 2007, IEEE Transactions on Image Processing.

[20]  D. Lu,et al.  Change detection techniques , 2004 .

[21]  Dongmei Chen,et al.  Change detection from remotely sensed images: From pixel-based to object-based approaches , 2013 .

[22]  Liangpei Zhang,et al.  Automatic Labelling and Selection of Training Samples for High-Resolution Remote Sensing Image Classification over Urban Areas , 2015, Remote. Sens..

[23]  Liangpei Zhang,et al.  A Nonlinear Multiple Feature Learning Classifier for Hyperspectral Images With Limited Training Samples , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[24]  S. Linden,et al.  Support vector regression and synthetically mixed training data for quantifying urban land cover , 2013 .

[25]  Annemarie Schneider,et al.  Monitoring land cover change in urban and peri-urban areas using dense time stacks of Landsat satellite data and a data mining approach , 2012 .

[26]  Zhaohui Xue,et al.  Phenology-Driven Land Cover Classification and Trend Analysis Based on Long-term Remote Sensing Image Series , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[27]  Huiping Liu,et al.  Trajectory-based detection of urban expansion using Landsat time series , 2014 .

[28]  Eyal Ben-Dor,et al.  Quality assessment of several methods to recover surface reflectance using synthetic imaging spectroscopy data , 2004 .

[29]  Richard G. Lathrop,et al.  Urban change detection based on an artificial neural network , 2002 .

[30]  W. Cohen,et al.  North American forest disturbance mapped from a decadal Landsat record , 2008 .

[31]  Jon Atli Benediktsson,et al.  A Novel Unsupervised Sample Collection Method for Urban Land-Cover Mapping Using Landsat Imagery , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[32]  Annemarie Schneider,et al.  Expansion and growth in Chinese cities, 1978–2010 , 2014 .

[33]  Mikhail F. Kanevski,et al.  A Survey of Active Learning Algorithms for Supervised Remote Sensing Image Classification , 2011, IEEE Journal of Selected Topics in Signal Processing.

[34]  Yun Zhang,et al.  Multispectral change detection using multivariate Kullback-Leibler distance , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[35]  Chun-Shin Lin,et al.  Seam-line determination for image mosaicking: A technique minimizing the maximum local mismatch and the global cost , 2010 .

[36]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[37]  P. Gong,et al.  A 30-year (1984–2013) record of annual urban dynamics of Beijing City derived from Landsat data , 2015 .

[38]  David P. Roy,et al.  Landsat-8 and Sentinel-2 burned area mapping - A combined sensor multi-temporal change detection approach , 2019, Remote Sensing of Environment.

[39]  Xin Huang,et al.  A novel co-training approach for urban land cover mapping with unclear Landsat time series imagery , 2018, Remote Sensing of Environment.