Multimedia Tools and Applications

, Volume 77, Issue 23, pp 30487–30504 | Cite as

Analysis of rain fall and the temperature of Coimbatore District using land use and land cover change detection by image segmentation

  • V. KalpanaEmail author


Image segmentation is a process has done for the classification of high resolution remote sensing images in the present research work. The segmentation results are capable of influencing the subsequent process effects. An image can be partitioned into a number of disjoint segments which is used to represent the image structures. It is found that it is more compact to represent an image and the low level and high structures can be combined. There are different types of methods to segment an image namely, threshold-based, edge-based and region-based. Region growing approach is image segmentation methods in which the neighboring pixels are examined and merged with the class region in case of no edges are detected. The iteration is done for every pixel boundary. Unlike gradient and Laplacian methods, the edges of the region are found by the region growing and it is perfectly their region. The images are determined by the LANDSAT TM satellite data. The remote sensing technique is used for collecting information about the Coimbatore district. The sensed data is a key to many diverse applications. The contribution of this work for Coimbatore district is to find the change of the Land used and Land covered in the entire region and also to find the changes in the green lands, vegetation and Land surface utilized for urban area. The neighboring regions are taken into account and the similarities are checked in the growing process. No single region is allowed to dominate the entire proceedings. A certain number of regions are allowed to grow at a time. Comparable regions will gradually combine into expanding regions. The Control of these methods may be quite complicated but efficient methods have been developed. The directions of growing pixels are easy and efficient to implement on parallel computers. The threshold-based segmentation is completely depending on the gray level images which regards the reflectivity of the featured images. It determines a threshold based on brightness of the ground objects. It is purely from the image background. But it is rapid and its uncertainty is significant. It is not convenient to process multi-spectral images.


Land use Land cover Region growing Segmentation Remote sensing Multi temporal 



The author is grateful to the Director, IRS, Anna University, and Chennai who helped to extract remotely sensed data for Coimbatore District, in this excellent work. The authors are thankful to Anna university Research Centre delegates who helped them in all this work since 2007.

She is grateful to Dr.K.Thanushkodi, Director, Akshaya College of Engineering and Technology, Coimbatore for his continuous encouragement in publishing this research work. She is also pleased to thank her husband Mr.G.Velmurugan and her kids V.A.Neya and V.Sornamugash for their countless encouragement and help in the work.


  1. 1.
    Baby Kalpana Y, Thanushkodi K, Sharrath M (2013) Change detection and classification using remotely sensed data for Coimbatore district. Jokull J 1(5):32–36Google Scholar
  2. 2.
    Bhatta B (2011) Remote sensing and GIS, 2nd ed. Oxford University Press, OxfordGoogle Scholar
  3. 3.
    Bins LS, Fonseca LMG (1996) Satellite imagery segmentation: a region growing algorithm. Anais VIII simpedio Brasilerio de sensoriamento Remoto, Brasil, 14–19 april 1996. Journal of Neuroscience, Psychology, and Economics® (JNPE), Brasil, pp 677–680Google Scholar
  4. 4.
    Canny JF (1986) Edge detection. IEEE Trans Pattern Anal Mach Intell Arch 8(6):679–698CrossRefGoogle Scholar
  5. 5.
    Castleman KR (2010) Digital image processing, 4th ed. Pearson Education Inc., Upper Saddle RiverGoogle Scholar
  6. 6.
    Coimbatore Corporation - SHB002 Coimbatore Corporation. Retrieved 2009-09-23
  7. 7.
    Historical Weather for Coimbatore, India Weather base. Retrieved 2009–09-23
  8. 8.
    John Cipar RL, Wood TC (2007) Testing an automated unsupervised classification algorithm with diverse land covers. IEEE Trans 1:2589–2592Google Scholar
  9. 9.
    Kamdi S, Krishna RK (2012) Image segmentation and region growing algorithm. (ISO 9001:2008 Certified International Journal, ISSN 2249-6343(Online)). Int J Comput Technol Electron Eng 2(1):103–107Google Scholar
  10. 10.
    Mathivanan SS (2012) High spatial resolution remote sensing image segmentation using marker based watershed algorithm. J Acad Indus Res J Acad Ind Res (JAIR) 1(15):257–260Google Scholar
  11. 11.
    O’Hara CG, King JS, Cartwright JH, King RL (2003) Multitemporal land use and land cover classification of urbanize areas within sensitive coastal environments. IEEE Transaction on Unsupervised Classification 41(9):2005–2017Google Scholar
  12. 12.
    Seasonal rainfall forecast for Southwest monsoon (2016)
  13. 13.
    Vincent S (1991) Watershed in digital spaces: an efficient algorithm based on immersion simulations. IEEE Trans Pattern Anal Mach Intell 13(6):583–589CrossRefGoogle Scholar
  14. 14.
    Yokoyama S, Kurosu T, Chiba C, Fujita M (2001) Land use classification with textural analysis and the aggregation technique using multi-temporal JERS-1 L -band SAR images. Int J Remote Sens 22(4):595–613CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of CSEP.A. College of Engineering and TechnologyPollachiIndia

Personalised recommendations