1 / 19

Lorenzo Bruzzone Francesca Bovolo

A SEMANTIC-BASED MULTILEVEL APPROACH TO CHANGE DETECTION IN VERY HIGH GEOMETRICAL RESOLUTION MULTITEMPORAL IMAGES. Lorenzo Bruzzone Francesca Bovolo. E-mail: lorenzo.bruzzone@ing.unitn.it Web page: http://rslab.disi.unitn.it. Outline. Introduction on change detection in VHR images. 1.

blakethomas
Download Presentation

Lorenzo Bruzzone Francesca Bovolo

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. A SEMANTIC-BASED MULTILEVEL APPROACH TO CHANGE DETECTION IN VERY HIGH GEOMETRICAL RESOLUTION MULTITEMPORAL IMAGES Lorenzo Bruzzone Francesca Bovolo E-mail: lorenzo.bruzzone@ing.unitn.it Web page: http://rslab.disi.unitn.it

  2. Outline Introduction on change detection in VHR images 1 General approach to change detection in VHR images 2 Illustration on the use of the approach for the solution of a specific change detection problem 3 4 Experimental results Conclusion 5 Lorenzo Bruzzone, Francesca Bovolo

  3. Introduction: Change Detection in VHR Images • Main assumption: unsupervised change-detection techniques generally assume that multitemporal images are similar to each other except for the presence of changes occurred on the ground. Problems: This assumption is seldom satisfied in VHR images due to: • the complexity of the objects present in the scene (which may show different spectral behaviors at two different dates even if their semantic meaning does not change); • the differences in the acquisition conditions (e.g., sensor acquisition geometry, atmospheric and sunlight conditions, etc.). Lorenzo Bruzzone, Francesca Bovolo

  4. Introduction: Change Detection in VHR Images October 2005 July 2006 Quickbird images acquired on a portion of the city of Trento (Italy) Lorenzo Bruzzone, Francesca Bovolo 4

  5. Aim of the Work • We propose a general top-down approach to the definition of the architecture of change detection methods for multitemporal VHR images. • The proposed approach: • explicitly models the presence of different radiometricchanges on the basis of the properties of the considered images • extracts the semantic meaning of changes; • identifies changes of interest with strategies designed on the basis of the specific application; • exploits the intrinsic multiscale properties of the objects and the high spatial correlation between pixels in a neighborhood. Lorenzo Bruzzone, Francesca Bovolo 5

  6. Proposed Approach: Architecture Design Multitemporal data set Auxiliary information Identification of the tree of radiometric changes Detection of all radiometric changes Selection of the strategy for detecting changes of interest Direct extraction of changes of interest Differential extraction of changes of interest by cancellation Detection of the changes of interest Refined detection of the radiometric change of interest Change detection map Lorenzo Bruzzone, Francesca Bovolo 6

  7. Identification of the Tree of Radiometric Changes Radiometric Changes(Wrad) Changes due to acquisition conditions (WAcq) Changes occurred on the ground (WGrd) Differences in acquisition system (WSys) Natural disasters (WDis) Differences in atmospheric conditions (WAtm) Anthropic activity (WAnt) Vegetation Phenology (Wveg) Type of sensor Sensor acquisition mode Environmental conditions (WEnv) Sensor view angle Seasonal effects Lorenzo Bruzzone, Francesca Bovolo 7

  8. Proposed Approach: Architecture Design Multitemporal data set Auxiliary information Identification of the tree of radiometric changes Change Vector Analysis, Context-sensitive techniques, etc. Detection of all radiometric changes Selection of the strategy for detecting changes of interest Direct extraction of changes of interest Differential extraction of changes of interest by cancellation Detection of the changes of interest Refined detection of the radiometric change of interest Change detection map Lorenzo Bruzzone, Francesca Bovolo 8

  9. Detection of Changes of Interest Differential detection by cancellation Direct detection of changes of interest X1 X1 X2 X2 Detection of radiometric changes Detection of change of interest 1 Detection of change of interest K Non-relevant change 1 Non-relevant change 2 Non-relevant change N - - - + + + + + Refined detection of the radiometric change of interest Map of changes Map of changes Lorenzo Bruzzone, Francesca Bovolo 9

  10. Multilevel Architecture: Semantic of Changes Object Meta-level (o) O1 O2 O j=1,…,Jo Classification map, object map,… Primitive Meta-level (p) P1 P2 P Meta-levels fusion j=1,…,Jp Geometric or statistic primitives Map of a specific Radiometric change Pixel Meta-level (px) X1 X2 D j=1,…,Jpx Pixel radiometry Lorenzo Bruzzone, Francesca Bovolo

  11. Data Set Description Study area: South part of Trento (Italy). Multitemporal data set: portion (380×430 pixels) of two images acquired by the Quickbird satellite in October 2004 and July 2006. Causes of Change:changes on the ground, seasonal changes, registration noise. October 2004 July 2006 Reference Map

  12. Proposed Approach: Architecture Design Multitemporal data set Auxiliary information Identification of the tree of radiometric changes Change Vector Analysis, Context-sensitive techniques, etc. Detection of all radiometric changes Selection of the strategy for detecting changes of interest Direct extraction of changes of interest Differential extraction of changes of interest by cancellation Detection of the changes of interest Refined detection of the radiometric change of interest Change detection map Lorenzo Bruzzone, Francesca Bovolo 12

  13. Identification of the Tree of Radiometric Changes WRad WSys WGrd wsh wrn WVeg WAnt Shadow changes wgl wat wb Registration noise Grassland Apple trees New buildings Lorenzo Bruzzone, Francesca Bovolo

  14. Changes Tree and Detection Strategy Differential detection by cancellation Identification of the tree of radiometric changes X1 X2 Detection of radiometric Changes (CVA) WRad WSys WGrd Detection of wsh Detection of wrn - - + + wsh wrn Refined detection of WGrd Shadow changes Registration noise Map of changes Lorenzo Bruzzone, Francesca Bovolo

  15. Multilevel Representation of Radiometric Changes S. Marchesi, F. Bovolo, L. Bruzzone, “A Context-Sensitive Technique Robust to Registration Noise for Change Detection in VHR Multispectral Images”, IEEE Transactions on Image Processing, Vol. 19, pp. 1877-1889, 2010. Registration noise map Primitive Meta-level (p) Segmentation map F. Bovolo, “A Multilevel Parcel-Based Approach to Change Detection in Very High Resolution Multitemporal Images,” IEEE Geoscience and Remote Sensing Letters, Vol. 6, No. 1, pp. 33-37, January 2009. Parcel map Shadow Index V. J. D. Tsai, "A comparative study on shadow compensation of color aerial images in invariant color models," IEEE Trans. Geosci. Remote Sens., vol. 44, pp. 1661-1671, 2006. Shadow change index Pixel Meta-level (px) X1 X2 L. Bruzzone and D. Fernández-Prieto, "Automatic Analysis of the Difference Image for Unsupervised Change detection," IEEE Trans. Geosci. Rem. Sens., vol. 38, pp. 1170-1182, 2000. Magnitude of multispectral change vectors Image radiometry Lorenzo Bruzzone, Francesca Bovolo 15

  16. Proposed Approach: Block Scheme Multiscale analysis for wrn detection Shadow detection Comparison wsh detection - - Shadowchange index + Shadowindex CVA Wrad detection Change-detection map X1 Magnitudeof multispectral changevectors W={Wnc, WGrd} Parcel detection X2 Lorenzo Bruzzone, Francesca Bovolo 16

  17. Experimental Results Overall change detection accuracy (%) 95 93.91 91.56 90 90.86 85 80 CVA Pixel-based CVA parcel-based Proposed method Marzo 2011

  18. Experimental Results October 2005 July 2006 Reference Map Change Detection map CVA parcel based Change detection map Proposed approach Marzo 2011

  19. Conclusion • We presented a general top-down approach to the definition of the architecture of change detection methods for multitemporal VHR images. • The main concepts exploited for the definition of the change detection architecture are: • Modeling the types of radiometric changes expected between images; • Extracting the semantic meaning from radiometric changes. • The approach proposed includes: • Direct detection of changes of interest or differential cancellation of uninteresting radiometric changes; • Multilevel and context-sensitive techniques; • Iterative strategy. • The approach has been successfully applied to the definition of aneffective architecture for change detection between Quickbird images in different application scenarios. Lorenzo Bruzzone, Francesca Bovolo

More Related