Open Access

Performance evaluation of brain tumor detection using watershed Segmentation and thresholding

Shruti Mishra

Shruti Mishra

School of Electrical Engineering, Vellore Institute of TechnologyVellore, India
Search for this author on
Sciendo|Google Scholar
Mishra, Shruti
, 
Noyonika Roy

Noyonika Roy

School of Electrical Engineering, Vellore Institute of TechnologyVellore, India
Search for this author on
Sciendo|Google Scholar
Roy, Noyonika
, 
Meghana Bapat

Meghana Bapat

School of Electrical Engineering, Vellore Institute of TechnologyVellore, India
Search for this author on
Sciendo|Google Scholar
Bapat, Meghana
 and 
Abhishek Gudipalli

Abhishek Gudipalli

School of Electrical Engineering, Vellore Institute of TechnologyVellore, India
Search for this author on
Sciendo|Google Scholar
Gudipalli, Abhishek
  
Nov 24, 2021
Performance evaluation of brain tumor detection using watershed Segmentation and thresholding's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Nov 24, 2021
Page range: 1 - 12
Received: Aug 20, 2021
DOI: https://doi.org/10.21307/ijssis-2021-020
Keywords
© 2021 Shruti Mishra et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1:

Proposed methodology.
Proposed methodology.

Figure 2:

Pre-processing techniques.
Pre-processing techniques.

Figure 3:

Pre-processing phase 1. (A) Grayscale image, (B) Image after median filter, (C) Image after smoothening, (D) Image after top-hat filter.
Pre-processing phase 1. (A) Grayscale image, (B) Image after median filter, (C) Image after smoothening, (D) Image after top-hat filter.

Figure 4:

Pre-processing phase 2. (A) Contrast enhancement, (B) Image after binarization, (C) Image after morphological erosion.
Pre-processing phase 2. (A) Contrast enhancement, (B) Image after binarization, (C) Image after morphological erosion.

Figure 5:

Processing (A) Thresholding, (B) Complement of image, (C) Image after watershed segmentation.
Processing (A) Thresholding, (B) Complement of image, (C) Image after watershed segmentation.

Figure 6:

Processing and post-processing techniques.
Processing and post-processing techniques.

Figure 7:

Post-processing (A) Image with four different coloured clusters superimposed after K-Means clustering and (B) Image after MSER detection.
Post-processing (A) Image with four different coloured clusters superimposed after K-Means clustering and (B) Image after MSER detection.

Figure 8:

Input image.
Input image.

Figure 9:

Output image.
Output image.

Comparative analysis_

Different approaches
ParametersProposed algorithm(Tarhini and Reda, 2020)(Oo and Aung Soe, 2014)
PSNR18.45379.7379518.97545
SSIM0.795450.18330.29315
FSIM0.925830.44390.75115
Entropy5.81220.911854.8462
NAE0.3680.99410.28255
NCC0.88010.375850.9122

Statistical analysis_

Parameter
ImagePSNRSSIMFISMEntropyNAENCC
Image 114.5480.7090.9616.2750.4490.875
Image 214.6940.6590.9396.0330.5010.804
Image 318.0210.6120.885.3010.4360.855
Image 414.260.7250.86666.1590.4630.775
Image 515.7750.7080.8814.80.6180.712
Image 619.6240.8090.8765.3370.2880.906
Image 718.9610.7480.9416.3760.30.948
Image 829.6730.9340.964.4730.080.991
Image 919.890.8860.9396.510.2710.931
Image 1017.9310.8060.9376.4930.3440.846
Language:
English
Publication timeframe:
1 times per year
Journal Subjects:
Engineering, Introductions and Overviews, Engineering, other
Journal RSS Feed