For decades even with the advancement of scanning and evolving imagery systems, the detection of tumors present in the brain has become increasingly challenging and is still one of the most sensitive areas in the field of medicine. In this research work, a enhanced model is proposed in which I have employed a combination of two algorithms. The first algorithm is the K Means Clustering which is a method of vector quantization. The second is the Fuzzy C-means clustering in which each data point can be attributed to more than one cluster. When used together the algorithms give enhanced performance and is called the TKFCM algorithm. Here in this paper, first in order to initialize the segmentation the k-means algorithm is used and the template is selected based on the Gray level of the image. Then the membership that is updated is primarily determined using the distance from the centroid of the cluster to the data points of the cluster using the FCM algorithm. Next, the improved FCM clustering algorithm is fundamentally used to detect the position of the tumor based on the function that is obtained by different features that include energy, contrast, entropy, homogeneity, dissimilarity and also the correlation. The results of the paper indicate that the algorithm is capable of detecting tissues that are abnormal in a improved time window.