Home / Current Issue / Paper 1703413
Detection of Lung Tumor Using Watershed Transformation Algorithm-based Image Enhancement
Subject area: Science,Engineering and Technology · Area of research: Image Processing
Abstract
Lung tumor detection and classification is one of the challenging tasks in medical image processing because of large variation in density, size, location of tumor and low contrast in images. The accuracy of such segmentation system should be high because it directly affects the mortality of humans. Existing classifier yields poor accuracy when volume of input increases. Hence it is more important to derive an algorithm with higher accuracy in lung tumor detection. In the proposed system, the image obtained from CT scan is processed using median filtering to remove noise and artifacts caused during transmission and other environmental facts. Then the filtered image is segmented using watershed segmentation algorithm and pixels are grouped to extract useful information for analysis. Performance of this algorithm is evaluated based on various metrics and value of bit error rate is calculated about 0.0153%. The result of this segmentation algorithm is better than the existing algorithms and also it preserves the edges of an image with minimum error rate.
Keywords
Lung, CT scan, Lung Tumour, Watershed transformation algorithm and Performance
References
[1] Bin-Habtoor, Abdulaziz Saleh Yeslem, and Salem Saleh Al-amri. "Removal speckle noise from medical image using image processing techniques." International Journal of Computer Science and Information Technologies 7, no. 1 (2016): 375-377.
[2] Sarah, Benmazou, Layachi Soumia, and Hayet Farida Merouani. "Segmentation of Images based cellular automata-reactive agent implemented in Netlogo platform." International Journal of Computer Applications 975 (2012): 8887.
[3] Divya, K. A., and K. I. Roshna. "A survey on various image enhancement algorithms for naturalness preservation." International Journal of Computer Science and Information Technologies 6, no. 3 (2015): 2043-2045.
[4] Patil, Dinesh D., and Sonal G. Deore. "Medical image segmentation: a review." International Journal of Computer Science and Mobile Computing 2, no. 1 (2013): 22-27.
[5] Nithila, Ezhil E., and S. S. Kumar. "Segmentation of lung nodule in CT data using active contour model and Fuzzy C-mean clustering." Alexandria Engineering Journal 55, no. 3 (2016): 2583-2588.
[6] Polat, Huseyin, and Homay Danaei Mehr. "Classification of pulmonary CT images by using hybrid 3D-deep convolutional neural network architecture." Applied Sciences 9, no. 5 (2019): 940.
[7] Na'am, Jufriadif. "Edge detection on objects of medical image with enhancement multiple morphological gradient method." In 2017 4th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), pp. 1-7. IEEE, 2017.
How to cite this paper
@article{1703413,
author = {Arun B. Mathews, Krishna Prasad K},
title = {Detection of Lung Tumor Using Watershed Transformation Algorithm-based Image Enhancement},
journal = {Iconic Research And Engineering Journals},
year = {2022},
volume = {5},
number = {11},
pages = {34-38},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1703413.pdf},
abstract = {Lung tumor detection and classification is one of the challenging tasks in medical image processing because of large variation in density, size, location of tumor and low contrast in images. The accuracy of such segmentation system should be high because it directly affects the mortality of humans. Existing classifier yields poor accuracy when volume of input increases. Hence it is more important to derive an algorithm with higher accuracy in lung tumor detection. In the proposed system, the image obtained from CT scan is processed using median filtering to remove noise and artifacts caused during transmission and other environmental facts. Then the filtered image is segmented using watershed segmentation algorithm and pixels are grouped to extract useful information for analysis. Performance of this algorithm is evaluated based on various metrics and value of bit error rate is calculated about 0.0153%. The result of this segmentation algorithm is better than the existing algorithms and also it preserves the edges of an image with minimum error rate.},
keywords = {Lung, CT scan, Lung Tumour, Watershed transformation algorithm and Performance},
month = {May},
}