Détection de changements structurels sur des images satellite haute résolution
Application en milieu forestier
OPEN ACCESS
Brutal and massive environmental changes, generally affecting large areas, have to be localized as rapidly as possible in order to manage the immediate impact of this type of events on ecosystems and prevent related risks. Therefore, it is necessary to develop efficient methods for change mapping. A quasi-unsupervised region-based method for change detection in high resolution satellite images is proposed. An automatic feature selection optimizes image segmentation and classification via an original calibration-like procedure. A binary classification enables then to separate altered from intact areas thanks to a new spatio-temporal descriptor based on the level of fragmentation of obtained regions. Both segmentation and classification involve a mean shift procedure. The method was assessed on forest environment using a Formosat-2 multispectral satellite image pair acquired before and after a major storm to identify and map the damages.
RÉSUMÉ
Les changements environnementaux brutaux et massifs, qui affectent g´en´eralement de grandes surfaces, doivent ^etre localis´es le plus rapidement possible pour g´erer l’impact imm´ediat de ce type d’´ev`enements sur les ´ecosyst`emes et pr´evenir les risques associ´es. Il est donc n´ecessaire de d´evelopper des m´ethodes permettant d’´etablir efficacement une carte des changements. Dans cette optique, une approche r´egion quasi non supervis´ee de d´etection de changements sur des images satellite `a haute r´esolution spatiale est propos´ee. Un proc´ed´e innovant de s´election automatique d’attributs, inspir´e des proc´edures de calibrage, optimise la segmentation et la classification. Un nouveau descripteur spatio-temporel, bas´e sur le taux de fragmentation des r´egions d´etect´ees, permet alors de r´ealiser une classification binaire des changements en zones intactes et alt´er´ees. Cette m´ethode passe par des ´etapes de segmentation et de classification mean shift. L’approche a ´et´e ´evalu´ee en milieu forestier sur un couple d’images satellite multispectrales Formosat-2 acquises avant et apr`es une temp^ete majeure pour reconna^ıtre et cartographier les d´eg^ats.
mapping, region classification, change detection, multispectral image, segmentation, feature selection.
MOTS-CLÉS
cartographie, classification de r´egions, d´etection de changements, image multispectrale, segmentation, s´election d’attributs
Blaschke T. (2010). Object based image analysis for remote sensing. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 65, no 1, p. 2 - 16.
Bo S., Ding L., Li H., Di F., Zhu C. (2009). Mean shift-based clustering analysis of multispectral remote sensing imagery. International Journal of Remote Sensing, vol. 30, no 4, p. 817{827.
Boukir S., Jones S., Reinke K. (2012). Fast mean-shift based classification of very high resolution images. application to forest cover mapping. In Isprs’2012, xxii international society for photogrammetry & remote sensing congress, isprs annals of photogrammetry, remote sensing and the spatial information sciences, melbourne, australia, vol. I-7, p. 111-116.
Boukir S., Orny C., Chehata N., Guyon D. (2011). Classification multitemporelle des deg^ats de temp^ete en for^et sur des images satellite haute r´esolution. In 23`eme colloque gretsi - traitement du signal et des images, bordeaux.
Bruzzone L., Prieto D. (2000). Automatic analysis of the difference image for unsupervised change detection. IEEE Trans. on Geoscience and Remote Sensing, vol. 38, no 3, p. 1171-1182.
Chehata N., Orny C., Boukir S., Guyon D. (2011). Object-based forest change detection using high resolution satellite images. In Pia’2011, isprs photogrammetric image analysis conference, international archives of photogrammetry, remote sensing and spatial information sciences, munich, germany, vol. 38, p. 49-54.
Cheng H., Jiang X., Sun Y., Wang J. (2001). Color image segmentation : advances and prospects. Pattern Recognition, vol. 34, no 12, p. 2259-2281.
Clandillon S., Yesou H., Meyer C. (2003). Benefits of spot 5 hr and vhr data for forest management and windfall damage mapping. In Ieee international geoscience and remote sensing symposium, igarss ’03, vol. 5, p. 3079 - 3081.
Comaniciu D., Meer P. (2002). Mean shift : A robust approach toward feature space analysis. IEEE Transactions on pattern analysis and machine intelligence, vol. 24, no 5, p. 603{619.
Conchedda G., Durieux L., Mayaux P. (2008). An object-based method for mapping and change analysis in mangrove ecosystems. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 63, no 5, p. 578{589. (Theme Issue: Remote Sensing of the Coastal Ecosystems)
Coppin P., Jonckheere I., Nackaerts K., Muys B., Lambin E. (2004). Digital change detection methods in ecosystem monitoring : a review. Int. Jour. of Remote Sensing, vol. 25, no 9, p. 1565-1596.
Descl´ee B., Bogaert P., Defourny P. (2006). Forest change detection by statistical object-based method. Remote Sensing of Environment, vol. 102, no 1-2, p. 1{11. Duda R., Hart P., Stork D. (2001). Pattern classification (2nd ed.). WileyInterscience.
Franklin S., Waring R., McCreight R., Cohen W., Fiorella M. (1995). Aerial and satellite sensor detection and classification of western spruce budworm defoliation in a subalpine forest. Canadian Journal of Remote Sensing, vol. 21, p. 299 - 308.
Fraser R., Abuelgasim A., Latifovic R. (2005). A method for detecting large-scale forest cover change using coarse spatial resolution imagery. Remote Sensing of Environment, vol. 95, no 4, p. 414{427.
Fukunaga K., Hostetler L. (1975). The estimation of the gradient of a density function, with applications in pattern recognition. IEEE Transactions on Information Theory, vol. 1, p. 32-40.
Fuller R., Smith G., Devereux B. (2003). The characterisation and measurement of land cover change through remote sensing : problems in operational applications. Int. Jour. of Applied Earth Observation and Geoinformation, vol. 4, p. 243-253.
Gonzales R. C., Woods R. E. (2008). Digital image processing (P. I. Edition, Ed.). Pearson Education Inc.
Guyon D., Courrier G., Berbigier P. (2001). Sensitivity analysis of coniferous forest reflectance with canopy structure and undergrowth characteristics from satellite data and modelling (case study: Landes maritime pine forest). In C. Ed (Ed.), 8th international symposium " physical measurements and signatures in remote sensing", p. 387-392. Aussois, France.
Hagolle O., Huc M., Pascual D. V., Dedieu G. (2010). A multi-temporal method for cloud detection, applied to formosat-2, venµs, landsat and sentinel-2 images. Remote Sensing of Environment, vol. 114, p. 1747-1755.
Haralick R., Shanmugan K., Dinstein I. (1973). Texture features for image classification. IEEE Transactions on Systems, Man and Cybernetics, vol. 3, p. 610-621.
Huang C., Song K., Kim S., Townshend J., Davis P., Masek J. G. et al. (2008). Use of a dark object concept and support vector machines to automate forest cover change analysis. Remote Sensing of Environment, vol. 112, no 3, p. 970{985.
Im J., Jensen J. (2005). A change detection model based on neighborhood correlation image analysis and decision tree classification. Remote Sensing of Environment, vol. 99, no 3, p. 326{340.
Johansen K., Arroyo L., Phinn S., Witte C., Kauth R., Thomas G. (2010). Comparison of geo-object based and pixel-based change detection of riparian environments using high spatial resolution multi-spectral imagery. Photogrammetric Engineering and Remote Sensing, vol. 76, no 2, p. 123{136.
Johansen K., Coops N., Gergel S., Stange Y. (2007). Application of high spatial resolution satellite imagery for riparian and forest ecosystem classification. Remote Sensing of Environment, vol. 110, no 1, p. 29{44.
Kennedy R., Townsend P., Gross J. E., Cohen W., Bolstad P., Wang Y. et al. (2009). Remote sensing change detection tools for natural resource managers: Understanding concepts and tradeoffs in the design of landscape monitoring projects. Remote Sensing of Environment, vol. 113, no 7, p. 1382-1396.
King D. J., Olthof I., Pellikka P. K. E., Seed E. D., Butson C. (2005). Modelling and mapping damage to forests from an ice storm using remote sensing and environmental data. Natural Hazards, vol. 35, p. 321 - 342.
Liu D., Song K., Townshend J., Gong P. (2008). Using local transition probability models in markov random fields for forest change detection. Remote Sensing of Environment, vol. 112, no 5, p. 2222{2231. (Earth Observations for Terrestrial Biodiversity and Ecosystems Special Issue)
Lunetta R., Johnson D., Lyon J., Crotwell J. (2004). Impacts of imagery temporal frequency on land cover change detection monitoring. Remote Sensing of Environment, vol. 89, p. 444-454.
Makela H., Pekkarinen A. (2001). Estimation of timber volume at the sample plot level by means of image segmentation and lansat tm imagery. Remote Sensing of Environment, vol. 77, p. 66-75.
Mallinis G., Koutsias N., Tsakiri-Strati M., Karteris M. (2008). Object-based classification using quickbird imagery for delineating forest vegetation polygons in a mediterranean test site. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 63, no 2, p. 237{250.
Melgani F., Bruzzone L. (2004). Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. on Geosciences and Remote Sensing, vol. 42, no 8, p. 1178-1790.
Meyer C., Geldreich P., Yesou H. (2001). Apports des donn´ees simul´ees spot5 pour l’´evaluation des d´eg^ats de temp^ete dans la for^et. cas de la for^et de haguenau (alsace, france). In Bulletin de la soci´et´e fran¸caise de photogramm´etrie et de t´el´ed´etection, vol. 164-165, p. 162-173.
Nielsen A., Conradsen K., Simpson J. (1998). Multivariate alteration detection (mad) and maf postprocessing in multispectral, bitemporal image data: New approaches to change detection studies. Remote Sensing of Environment, vol. 64, no 1, p. 1-19.
Olthof I., King D. J., Lautenschlager R. (2004). Mapping decidious forest ice storm damage using landsat and environmental data. Remote Sensing of Environment, vol. 89, p. 484-496.
Orny C., Chehata N., Boukir S., Guyon D. (2010). Characterization of maritime pine forest structure changes with vhr satellite imagery : application to the 24th january 2009 windfall damages cartography. Tech. Rep.. CNES.
Pacifici F., Frate F. D., Solimini C., Emery W. (2007). An innovative neural-net method to detect temporal changes in high-resolution optical satellite imagery. IEEE Trans. on Geosciences and Remote Sensing, vol. 45, no 1, p. 217-222.
Richards J. A., Jia X. (1999). Remote sensing digital image analysis: An introduction. Springer.
Ruiz L. A., Fdez-sarr´ıa A., Recio J. A. (2004). Texture feature extraction for classification of remote sensing data using wavelet decomposition: a comparative study. In International archives of photogrammetry and remote sensing. vol.xxxv , p. 1682{1750.
Schwarz M., Steinmeier C., Waser L. (2001). Detection of storm losses in alpine forest areas by different methodic approaches using high resolution satellite data. In Proceedings of the 21st earsel symposium: Observing our environment from space: New solutions for a new millenium, p. 251{257.
St-Louis V., Pidgeon A., Radeloff V., T.J.Hawbaker, Clayton M. (2006). Highresolution image texture as a predictor of bird species richness. Remote Sensing of Environment, vol. 105, p. 299-312.
Taud H., Couturier S., Carrillo-Rivera J. (2012). Remote sensing image classification by mean shift and colour quantization. In Image and signal processing for remote sensing xviii. proceedings of the spie, vol. 8537.
Trias-Sanz R., Stamon G., Louchet J. (2008). Using colour, texture, and hierarchical segmentation for high-resolution remote-sensing. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 63, p. 156-168.
Tuominen S., Pekkarinen A. (2005). Performance of different spectral and textural aerial photograph features in multi-source forest inventory. Remote Sensing of Environment, vol. 94, p. 256-268.
Wang Y., Soh Y., Shultz H. (2006). Individual tree crown segmentation in aerial forestry images by mean shift clustering and graph-based cluster merging. Computer Science and Network Security, vol. 6, no 11, p. 40 - 45.