Images have always been an attraction and our existence depends on them. Without images world would be an empty canvass. Our eyes capture thousands of images each day and our brain processes them. Interestingly we human can identify images in a micro second, a wink of an eye and we know what the object is, be it moving or static. Artificial intelligence is designed to let the computers think and behave like human beings, fuzzy logic is one of its important technique which has been used in the proposed thesis to segment and image. The proposed algorithm reads the image, pre-process it, then fuzzy rules are applied over it and finally de- fuzzification is carried over it to get the segmented image. The algorithm is compared with existing K-Mean and Modified K-Mean to access the viability of the proposed algorithm. The algorithm is tested for number of segments, segmented area, and time taken, it is observed that the proposed algorithm improves K-Mean, by 60%, 1.6%, 94% respectively and Modified K-Mean by 43%, 1.2%, 13.5% respectively. The results indicate that the proposed algorithm works better than the previous two algorithms. There is a marked improvement in number of segments maintaining the time taken. In this proposed worked has been overcome in MATLAB features.

Segmenting pixels images, Trim function, di-fuzzification, centroids methods, modified k means ,fuzzyfication

[1] Gurbinder Kaur , Balwinder Singh" Intensity Based image segmentation using wavelet analysis and clustering techniques" published in ijcse indian journol of computer science and engineering vol 2, no 3,2011.
[2] Navneet Kaur, Gagan Jindal, “A Survey Of K Means
Clustering With Modified Gradient Magnitude Region Growing Technique For Lesion Segmentation”, International Journal Of Innovations In Engineering And Technology, 2013.
[3] X. Cui, G. Yang, Y. Deng and S. Wu, “An Improved Image Segmentation Algorithm Based on the Watershed
Transform”, IEEE, pp. 428—431, 2014.
[4]Savita Agrawal et al “Survey of image segmentation techniques and color models” vol 5(3),2014, / (IJCSIT) International Journal of Computer Science and Information Technologies.
[5]Chenhang Zhou, Liwei Tian*,Hongwei Zhao, Kai Zhao,"A Method of Two-Dimensional Otsu Image Threshold Segmentation Based on Improved Firefly Algorithm"2015
[6] P.M.K. Prasad, D.Y.V. Prasad, G. Sasibhushana Rao Prof.,“Performance analysis of orthogonal and biorthogonal
wavelets for edge detection of xray images”, Procedia Computer Science, International Conference on Recent Trends in Computer Science & Engineering, Vol. 87, pp
116-121, 2016.
[6] Divya, Mr Pawan Kumar Mishra "Frequency Domain Digital Image Segmentation based on a Modified k Means" (ijircst) issn:2347-5552,vol 5,issue 4 ,2017.