Interpreting a tumor accurately in magnetic resonance imaging (MRI) brain tumor image is a challenging area of research. Image segmentation provides a solution to separate normal cells from the affected cells. It uses the discontinuity as well as similarity principle of the intensity values to separate the regions (which is similar to the edge detection process). In the proposed work, a MRI brain tumor (MBT) image is analyzed using fuzzy rule based approach (FRBA) to detect the highly active region (considered as tumor) and less active region. In this approach, a triangular membership function (TMF) is used to separate the intensity values into four different classes such as low, medium, high, and very high. Afterwards, a FRBA based on the standard RGB model is used to segment the MBT image into two parts such as red for highly active region and black for less active as well as other regions. The performance of this approach is analyzed using MATLAB R2015b. From the results, it is observed that the proposed approach performs better than Wu et al.  and Juang et al.  methods in detecting the tumor in terms of detection accuracy and outlier detection.