Dissertation > Excellent graduate degree dissertation topics show

The Study of the Oil Spills Detection and Classification Based on SAR Images

Author: LiQiong
Tutor: ChenJianPing
School: Chinese Geology University (Beijing)
Course: Earth Exploration and Information Technology
Keywords: Film extract Level set segmentation Texture Segmentation Neural network classifier Leakage of oil film Pollution film
CLC: TN957.52
Type: Master's thesis
Year: 2011
Downloads: 189
Quote: 0
Read: Download Dissertation

Abstract


Marine oil spill environmental hazards and property damage is very serious, its effective detection and monitoring of oil spill rapid response. Has developed a variety of film detection methods, including the use of SAR detection film has achieved very good results. This thesis is mainly to complete the of the SAR image oil slicks extraction and classification method. Radar detection of marine oil spill reflect sensitive targets on the ground surface roughness characteristics. After the oil spill occurred, spread to the sea of ??oil film damping capillary waves on the sea surface gravity, sea oil film than no film of the sea is smooth, so that smaller radar backscatter echo. By extracting image low-scattering region dark area on the image to complete the film extract. In this paper, a single threshold value, texture, level set image segmentation method to extract the oil film information. The level set method is popular in recent years, an edge detection method, which does not depend on the local information of the image edge detection fuzzy edge detection to obtain a very good detection results. Many ocean phenomena in the SAR images of the dark area of ??low backscatter after due SAR image scattered dark areas of multiple solutions, combined with the SAR image texture information intelligent recognition method using artificial neural networks to film classification, exclusion suspected oil film. Extracted based on GLCM method SAR image texture characteristic values, select the input mean, entropy, homogeneity, contrast, angular second moment five texture features indicators as BP neural network in the study used statistical methods to train the network ; then trained network be classified image each pixel classification; final simple evaluation of the effect of the classification. Characteristics of artificial neural network of intelligent recognition also makes use of this approach to the film classification will achieve better results. Paper last brief identification method the film extract business processes as well as pollution of oil film and leakage of oil film. SAR image based multi-source remote sensing images for the detection of oil slicks, and sea conditions, maritime transport, marine facilities, geological fault structure, gravity and magnetic anomalies, geochemical anomaly and other information-assisted, superimposed on time, not at the same time phase interprets the history of film to determine the contaminated film and leak oil film.

Related Dissertations

  1. Research on Recognition of Fabric Weave Pattern Based on Space-Frequency Domain,TS101.923
  2. Research and Implementation in Face Recognition Based on Multiple Classifiers,TP391.41
  3. Based on fractional Fourier transform texture image segmentation,TP391.41
  4. Research on Application of Texture Feature to Image Segmentation,TP391.41
  5. The Study on Multifractal Theory and Application in Image Processing,TP391.41
  6. Research on Target Detection in Remote Sensing Images,TP391.41
  7. Image Segmentation and Object Tracking of Computer-aided Diagnosis Outside the Anal Canal Disease,TP391.41
  8. A Multistage Neural Network System for Handwritten Numeral Recognition,TP391.41
  9. EMD-based medical image fusion algorithm,TP391.41
  10. Research on Algorithm of Video Object Segmentation,TP391.41
  11. Study on Difference of Gaussian Model-based Low Level Visual Feature Detection,TP391.4
  12. Geometric Flows Based Medical Image Segmentation Methods and Applications,TP391.41
  13. Research on Image Segmentation Based on Variational Methods and Fuzzy Clustering Algorithms,TP391.41
  14. Research of BP Network Signature Verification Using Statistics Features and Wavelet Features,TP309
  15. The Research of License Plate Character Recognition Based on BPNN,TP183
  16. Image Analysis and Encryption,TP391.41
  17. Fabric texture simulation study questions,TP391.9
  18. Research on Classifying ST Segment in ECG Using Modified Fuzzy ARTMAP,TP391.41
  19. Research and Realization of Metallographic Image Classification Based on Texture Feature,TP391.41
  20. Texture Segmentation Based on independent component analysis,TP391.41

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Radar > Radar equipment,radar > Radar receiving equipment > Data,image processing and admission
© 2012 www.DissertationTopic.Net  Mobile