Automating land cover mapping of Scotland using expert system and knowledge integration methods

Abstract Effective land cover mapping often requires the use of multiple data sources and data interpretation methods, particularly when no one data source or interpretation method provides sufficiently good results. Method-oriented approaches are often only effective for specific land cover class/data source combinations, and cannot be applied when different classification systems or data sources are required or available. Here we present a method, based on Endorsement Theory, of pooling evidence from multiple expert systems and spatial datasets to produce land cover maps. Individual ‘experts’ are trained to produce evidence for or against a class, with this evidence being categorised according to strength. An evidence integration rule set is applied to evidence lists to produce conclusions of different strength regarding individual classes, and the most likely class identified. The only expert system design implemented currently within the methodology is a neural network model, although the system has been designed to accept information from decision trees, fuzzy k-means and Bayesian statistics as well. We have used the technique to produce land cover maps of Scotland using three classification systems of varying complexity. Mapping accuracy varied between 52.6% for a map with 96 classes to 88.8% for a map with eight classes. The accuracy of the maps generated is higher than when individual datasets are used, showing that the evidence integration method applied is suitable for improving land cover mapping accuracy. We showed that imagery was not necessarily the most important data source for mapping where a large number of classes are used, and also showed that even data sources that produce low accuracy scores when used for mapping by themselves do improve the accuracy of maps produced using this integrative approach. Future work in developing the method is identified, including the inclusion of additional expert systems and improvement of the evidence integration, and evaluation carried out of the overall effectiveness of the approach.

[1]  D. Amarsaikhan,et al.  The integrated use of optical and InSAR data for urban land‐cover mapping , 2007 .

[2]  Paul R. Cohen,et al.  Heuristic reasoning about uncertainty: an artificial intelligence approach , 1984 .

[3]  T. R. Vijayaram,et al.  Materials handling technology and significance of expert systems to select appropriate handling equipments in engineering industries : A review , 2006 .

[4]  Hugh G. Lewis,et al.  Superresolution mapping using a hopfield neural network with fused images , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[5]  Rick L. Lawrence,et al.  Mapping prairie pothole communities with multitemporal Ikonos satellite imagery. , 2006 .

[6]  Shu-Hsien Liao,et al.  Expert system methodologies and applications - a decade review from 1995 to 2004 , 2005, Expert Syst. Appl..

[7]  H. Pereira,et al.  Towards the global monitoring of biodiversity change. , 2006, Trends in ecology & evolution.

[8]  Corinna Gries,et al.  Estimating vegetation cover in an urban environment based on Landsat ETM+ imagery: A case study in Phoenix, USA , 2007 .

[9]  G. G. Wright,et al.  Landsat TM spectral information to enhance the land cover of Scotland 1988 dataset , 1997 .

[10]  Jeffrey G. Masek,et al.  Landsat-7 ETM+ as an observatory for land cover: Initial radiometric and geometric comparisons with Landsat-5 Thematic Mapper , 2001 .

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

[12]  Willie Towers,et al.  Landscape-scale vegetation change in the Cairngorms, Scotland, 1946-1988: implications for land management , 1996 .

[13]  Yafit Cohen,et al.  Analysis of convergent evidence in an evidential reasoning knowledge-based classification , 2005 .

[14]  Mark A. Friedl,et al.  Using prior probabilities in decision-tree classification of remotely sensed data , 2002 .

[15]  R. Fournier,et al.  A comparison of four methods to map biomass from Landsat-TM and inventory data in western Newfoundland , 2006 .

[16]  P. Defourny,et al.  Retrieving forest structure variables based on image texture analysis and IKONOS-2 imagery , 2006 .

[17]  Alexis J. Comber,et al.  A comparison of Bayes', Dempster-Shafer and Endorsement theories for managing knowledge uncertainty in the context of land cover monitoring , 2004, Comput. Environ. Urban Syst..

[18]  Matt Aitkenhead,et al.  Improving land-cover classification using recognition threshold neural networks , 2007 .

[19]  Thuy Le Toan,et al.  Use of ENVISAT ASAR wide-swath mode data over Siberia for large area land cover mapping, parameter retrieval, and change detection , 2005, Proceedings. 2005 IEEE International Geoscience and Remote Sensing Symposium, 2005. IGARSS '05..

[20]  Alex B. McBratney,et al.  Digital terron mapping , 2005 .

[21]  D. Amarsaikhan,et al.  Data fusion and multisource image classification , 2004 .

[22]  Tatiana Mora Kuplich,et al.  Classifying regenerating forest stages in Amazônia using remotely sensed images and a neural network , 2006 .

[23]  H. Tian,et al.  China's changing landscape during the 1990s: Large‐scale land transformations estimated with satellite data , 2005 .

[24]  Richard A. Wadsworth,et al.  Setting Site Specific Critical Loads: An Approach using Endorsement Theory and Dempster–Shafer , 2007 .

[25]  S. Phinn,et al.  Analysis of multi-date MISR measurements for forest and woodland communities, Queensland, Australia , 2007 .

[26]  J. Lucio,et al.  Relationship between landscape visual attributes and spatial pattern indices: A test study in Mediterranean-climate landscapes , 2006 .

[27]  Alan H. Strahler,et al.  Global land cover mapping from MODIS: algorithms and early results , 2002 .

[28]  Hugh Kelley,et al.  Assessing the transition from deforestation to forest regrowth with an agent-based model of land cover change for south-central Indiana (USA) , 2008 .

[29]  David A. Clausi,et al.  Preserving boundaries for image texture segmentation using grey level co-occurring probabilities , 2006, Pattern Recognit..

[30]  C. Furlanello,et al.  Predicting habitat suitability with machine learning models: The potential area of Pinus sylvestris L. in the Iberian Peninsula , 2006 .

[31]  P. Englefield,et al.  Fine-resolution mapping of wildfire fuel types for Canada: Fuzzy logic modeling for an Alberta pilot area , 2006, Environmental monitoring and assessment.

[32]  Jianhua Gong,et al.  Genetic algorithm-based decision tree classifier for remote sensing mapping with SPOT-5 data in the HongShiMao watershed of the loess plateau, China , 2007, Neural Computing and Applications.

[33]  Mahta Moghaddam,et al.  Integration of radar and Landsat-derived foliage projected cover for woody regrowth mapping, Queensland, Australia , 2006 .

[34]  G. Hudson,et al.  Climatic soil moisture deficit - climate and soil data integration in a GIS , 1994 .

[35]  William L. Stefanov,et al.  Land use and land cover mapping from diverse data sources for an arid urban environments , 2006, Comput. Environ. Urban Syst..

[36]  Martin Herold,et al.  Indicators of Northern Eurasia's land‐cover change trends from SPOT‐VEGETATION time‐series analysis 1998–2005 , 2007 .

[37]  A. Gieske,et al.  Crop and land cover classification in Iran using Landsat 7 imagery , 2006 .

[38]  A. D. Gregorio,et al.  Land Cover Classification System (LCCS): Classification Concepts and User Manual , 2000 .

[39]  Yingtao Jiang,et al.  The internet-based knowledge acquisition and management method to construct large-scale distributed medical expert systems , 2004, Comput. Methods Programs Biomed..

[40]  Arthur P. Dempster,et al.  A Generalization of Bayesian Inference , 1968, Classic Works of the Dempster-Shafer Theory of Belief Functions.

[41]  Saro Lee,et al.  Landslide susceptibility mapping using GIS and the weight-of-evidence model , 2004, Int. J. Geogr. Inf. Sci..

[42]  Giles M. Foody,et al.  Training set size requirements for the classification of a specific class , 2006 .

[43]  John Trinder,et al.  Using the Dempster-Shafer method for the fusion of LIDAR data and multi-spectral images for building detection , 2005, Inf. Fusion.

[44]  Taskin Kavzoglu,et al.  Increasing the accuracy of neural network classification using refined training data , 2009, Environ. Model. Softw..

[45]  Mark Rounsevell,et al.  The limitations of spatial land use data in environmental analysis , 2006 .

[46]  Scott N. Miller,et al.  Assessing land cover change in Kenya's Mau Forest region using remotely sensed data , 2008 .

[47]  Paul R. Cohen,et al.  An Endorsement-based Plan Recognition Program , 1985, IJCAI.

[48]  D. Boyd,et al.  Updating topographic mapping in Great Britain using imagery from high-resolution satellite sensors , 2006 .

[49]  Liviu Florin V. Zoran,et al.  Multisensor image fusion for Romanian Black Sea coastal zone analysis , 2006, SPIE Remote Sensing.

[50]  G. Foody.  Thematic map comparison: Evaluating the statistical significance of differences in classification accuracy , 2004 .

[51]  Glenn Shafer,et al.  A Mathematical Theory of Evidence , 2020, A Mathematical Theory of Evidence.

[52]  A. N. R. Law,et al.  A system for monitoring land cover , 2003 .

[53]  Ronald J. Hall,et al.  Operational mapping of the land cover of the forested area of Canada with Landsat data: EOSD land cover program , 2003 .

[54]  A M MacDonald,et al.  The impact of climate change on the soil/moisture regime of Scottish mineral soils. , 1994, Environmental pollution.

[55]  Kent W. Hilbert.  Land Cover Change within the Grand Bay National Estuarine Research Reserve: 1974–2001 , 2006 .

[56]  Giles M. Foody,et al.  Mapping specific habitats from remotely sensed imagery: Support vector machine and support vector data description based classification of coastal saltmarsh habitats , 2007, Ecol. Informatics.

[57]  Heather McNairn,et al.  Multispectral indices and advanced classification techniques to detect percent residue cover over agricultural crops using Landsat data , 2006, SPIE Optics + Photonics.

[58]  Zhisong Pan,et al.  An integrated model of intrusion detection based on neural network and expert system , 2005, 17th IEEE International Conference on Tools with Artificial Intelligence (ICTAI'05).

[59]  Iain Brown,et al.  Modelling future landscape change on coastal floodplains using a rule-based GIS , 2006, Environ. Model. Softw..

[60]  E. Bork,et al.  Integrating LIDAR data and multispectral imagery for enhanced classification of rangeland vegetation: A meta analysis , 2007 .

[61]  Geoff Smith,et al.  The characterisation and measurement of land cover change through remote sensing: problems in operational applications? , 2003 .

[62]  Kenneth J. Mackin,et al.  Land Cover Classification from MODIS Satellite Data Using Probabilistically Optimal Ensemble of Artificial Neural Networks , 2006, KES.

[63]  Michele Meroni,et al.  Combining medium and coarse spatial resolution satellite data to improve the estimation of sub-pixel NDVI time series , 2008 .

[64]  R. Lawrence,et al.  Mapping wetlands and riparian areas using Landsat ETM+ imagery and decision-tree-based models , 2006, Wetlands.

[65]  Xiaohu Zhang,et al.  Determining Class Proportions Within a Pixel Using a New Mixed-Label Analysis Method , 2010, IEEE Transactions on Geoscience and Remote Sensing.