Abstract
Mixed pixels are a major problem in mapping land cover from remotely sensed imagery. Unfortunately, such imagery may be dominated by mixed pixels, and the conventional ‘hard’ image classification techniques used in mapping applications are unable to appropriately represent the land cover of mixed pixels. Fuzzy classification techniques can, however, accommodate the partial and multiple class membership of mixed pixels, and be used to derive an appropriate land cover representation. This is, however, only a partial solution to the mixed pixel problem in supervised image classification. It must be reognised that the land cover on the ground is fuzzy, at the scale of the pixel, and so it is inappropriate to use procedures designed for ‘hard’ data in the training and testing stages of the classification. Here an approach for land cover classification in which fuzziness is accommodated in all three stages of a supervised classification is presented. Attention focuses on the classification of airborne thematic mapper data with an artificial neural network. Mixed pixels could be accommodated in training the artificial neural network, since the desired output for each training pixel can be specified. A fuzzy land cover representation was derived by outputting the activation level of the network's output units. The activation level of each output unit was significantly correlated with the proportion of the area represented by a pixel which was covered with the class associated with the unit (r>0.88, significant at the 99% level of confidence). Finally, the distance between the fuzzy land cover classification derived from the artificial neural network and the fuzzy ground data was used to illustrate the accuracy of the land cover representation derived. The dangers of hardening the classification output and ground data sets to enable a conventional assessment of classification accuracy are also illustrated; the hardened data sets were over three times more distant from each other than the fuzzy data sets.
Similar content being viewed by others
References
Rhind D, Hudson R. Land Use. Methuen, London, 1980.
Estes JE, Mooneyhan DW. Of maps and myths. Photogrammetric Engineering and Remote Sensing 1994; 60: 517–524.
Townshend JRG. Land cover. Int J of Remote Sens 1992; 13: 1319–1328.
Herman JD, Waites JE, Ponitz RM, Etzler P. A temporal and spatial remote sensing study of a Michigan superfund site. Photogrammetric Engineering and Remote Sensing 1994: 60: 1007–1017.
Fuller RM, Groom GB, Jones AR. The land-cover map of Great Britain: An automated classification of Landsat Thematic Mapper Data. Photogrammetric Engineering and Remote Sensing 1994; 60: 553–562.
Stone TA, Schlesinger P, Houghton RA, Woodwell GM. A map of the vegetation of South America based on satellite imagery. Photogrammetric Engineering and Remote Sensing 1994; 60: 541–551.
DeFries RS, Townshend JRG. Global land cover: comparison of ground-based data sets to classification with AVHRR data. In: Foody GM, Curran PJ (eds.), Environmental Remote Sensing from Regional to Global Scales. Wiley, Chichester, 1994, pp. 84–110.
Conese, C, Maselli, F. Use of error matrices to improve area estimates with maximum likelihood classification procedures. Remote Sens of Environ 1992; 40: 113–124.
Benediktsson JA, Swain PH, Ersoy OK. Neural network approaches versus statistical methods in classification of multisource remote sensing data. IEEE Trans Geosci Remote Sens 1990; 28: 540–551.
Peddle DR, Foody GM, Zhang A, Franklin SE, LeDrew EF. Multisource image classification II: an empirical comparison of evidential reasoning and neural network approaches. Canadian J Remote Sens 1994; 20 (4): 397–408.
Wood TF, Foody GM. Analysis and representation of vegetation continua from Landsat Thematic Mapper data for lowland heaths. Int J Remote Sens 1989; 10: 181–191.
Campbell JB. Introduction to Remote Sensing. Guilford, New York, 1987.
Crapper PF. An estimate of the number of boundary cells in a mapped landscape coded to grid cells. Photogrammetric Engineering and Remote Sensing 1984; 50: 1497–1503.
Skoke D, Tucker C. Tropical deforestation and habitat fragmentation in the Amazon: satellite data from 1978 to 1988. Science 1993; 260: 1905–1910.
Wang F. Improving remote sensing image analysis through fuzzy information representation. Photogrammetric Engineering and Remote Sensing 1990; 56: 1163–1169.
Wang F. Fuzzy supervised classification of remote sensing images. IEEE Trans Geosci Remote Sens 1990; 28: 194–201.
Cannon RL, Dave JV, Bezdek JC, Trivedi MM. Segmentation of a thematic mapper image using the fuzzyc-means clustering algorithm. IEEE Trans Geosci Remote Sens 1986; 24: 400–408.
Foody GM, Campbell, NA, Trodd, NM, Wood TF. Derivation and applications of probabilistic measures of class membership from the maximum likelihood classification. Photogrammetric Engineering and Remote Sensing 1992; 58: 1335–1341.
Foody GM. Fuzzy modelling of vegetation from remotely sensed imagery. Ecological Modelling 1996; 85, 3–12.
Metzler MD, Cicone RC. Assessment of technologies for classification of mixed pixels. Proceedings of the Seventeenth International Symposium on Remote Sensing of Environment, vol 3, ERIM, Ann Arbor Michigan, 1983, pp. 1015–1021.
Arai K. A supervised Thematic Mapper classification with a purification of training samples. Int J Remote Sens 1992; 13: 2039–2049.
Schalkoff RJ. Pattern Recognition: Statistical, Structural and Neural Approaches. Wiley, New York, 1992.
Aleksander I, Morton H. An Introduction to Neural Computing. Chapman & Hall, London, 1990.
Davalo E, Naim P. Neural Networks. Macmillan, Basingstoke, 1991.
de Villiers J, Barnard E. Backpropagation neural nets with one and two hidden layers. IEEE Trans Neural Networks 1992; 4: 136–141.
Chauvin Y. Generalization performance of overtrained back-propagation networks, In Almeida LB, Wellekens CJ (eds). Neural Networks. Springer-Verlag, Berlin, 1990, pp. 46–55.
Rumelhart DE, Hinton GE, Williams RJ. Learning internal representation by error propagation. In: Rumelhart DE, McClelland JL (eds.), Parallel Distributed Processing: Explorations in the Microstructure of Cognition: MIT Press, Cambridge MA, 1986, pp. 318–362.
Rosenfield GH, Fitzpatrick-Lins K. A coefficient of agreement as a measure of thematic classification accuracy. Photogrammetric Engineering and Remote Sensing 1986; 52: 223–227.
Congalton RG. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens Environ 1991: 37: 35–46.
Cohen J. A coefficient of agreement for nominal scales. Educational and Psychological Measurement 1960; 20: 37–46.
Story M, Congalton RG. Accuracy assessment: a user's perspective. Photogrammetric Engineering and Remote Sensing 1986; 52: 397–399.
Congalton RG, Oderwald RG, Mead RA. Assessing Landsat classification accuracy using discrete multivariate analysis statistical techniques. Photogrammetric Engineering and Remote Sensing 1983; 49: 1671–1678.
Gong P, Howarth PJ. The use of structural information for improving land-cover classification accuracies at the rural-urban fringe. Photogrammetric Engineering and Remote Sensing 1990; 56: 67–73.
Cohen J. Weighted kappa. Psychological Bulletin 1968; 70: 213–220.
Foody GM, Trodd NM, Non-classificatory analysis and representation of heathland vegetation from remotely sensed imagery. GeoJournal 1993; 29: 343–350.
Curran PJ, Hay AM. The importance of measurement error for certain procedures in remote sensing at optical wavelengths. Photogrammetric Engineering and Remote Sensing 1986; 52: 229–241.
Curran PJ, Williamson HD. The accuracy of ground data used in remote sensing investigations. Int J Remote Sens 1985; 6: 1637–1651.
Bauer ME, Burk TE, Ek AR, Coppin PR, Lime SD, Walsh TA, Walters DK. Satellite inventory of Minnesota forest resources. Photogrammetric Engineering and Remote Sensing 1994; 60: 287–298.
Gopal S, Woodcock C. Theory and methods for accuracy assessment of thematic maps using fuzzy sets. Photogrammetric Engineering and Remote Sensing 1994; 60: 181–188.
Congalton RG. Accuracy assessment of remotely sensed data: future needs and directions. In Proceedings of Twelfth Pecora Symposium, American Society for Photogrammetry and Remote Sensing, Bethesda, 1994, pp. 385–388.
Finn JT. Use of the average mutual information index in evaluating classification error and consistency. Int J Geographical Infor Syst 1993; 7: 349–366.
Maselli F, Conese C, Petkov L. Use of probability entropy for the estimation and graphical representation of the accuracy of maximum likelihood classifications. ISPRS J Photogrammetry Remote Sens 1994; 49 (2): 13–20.
Klir GJ, Folger TA. Fuzzy Sets, Uncertainty and Information. Prentice-Hall International, London, 1988.
Foody GM. Approaches for the production and evaluation of fuzzy land cover classifications from remotely sensed data. Int J Remote Sens 1996; 17: 1317–1340.
Chang C-I, Chen K, Wang J, Althouse MLG. A relative entropy-based approach to image thresholding. Patt Recogn 1994; 27: 1275–1289.
Higashi M, Klir GJ. On the notion of distance representing information closeness: possibility and probability distributions. Int J General Syst 1983; 9: 103–115.
Kent JT, Mardia KV. Spatial classification using fuzzy membership models. IEEE Trans Pattern Anal Machine Intell 1988; 10: 659–671.
Goodchild MF. Integrating GIS and remote sensing for vegetation analysis and modelling: methodological issues. J Vegetation Sci 1994; 5: 615–626.
Townshend JRG. Agricultural land-cover discrimination using thematic mapper spectral bands. Int J Remote Sens 1984; 6: 681–698.
Foody GM, Cox DP. Sub-pixel land cover composition estimation using a linear mixture model and fuzzy membership functions. Int J Remote Sens 1994; 15: 619–631.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Foody, G.M. Fully fuzzy supervised classification of land cover from remotely sensed imagery with an artificial neural network. Neural Comput & Applic 5, 238–247 (1997). https://doi.org/10.1007/BF01424229
Issue Date:
DOI: https://doi.org/10.1007/BF01424229