Abstract: Brain tumors consist of several types of abnormal growth within the skull and the Central Nervous System (CNS). They result from abnormal and uncontrolled able cell division. Brain tumors are dangerous and when left untreated can result in loss of life. Some tumors, however does not lead to death, especially, the lipomas which are inherently unharmful. The potential risk associated with tumors depends on several factors such as the type of the tumuor its size and location as well as the condition of the tumor. This study proposes the use of pre-processing phase wave image filter size box algorithm for brain tumor imaging as this algorithm provides improved image quality devoid of noise. The suggested method contains three essential steps, the first step involves the use of a Genetic algorithm which has a zero tolerance for probability and flexibility and can also find near-optimal solutions. In the second step an entrained Particle Sswarm Optimization (PSO) technique was used to mechanically determine the mid-clustering the randomly collected data set. The third stage involved the merging of one and two regions of segmentation in the data set using marshalled to detect a tumor in the brain. The noise in the image was filtered using the Laplacian-Gaussian technique before edge detection. The Laplacian-Gaussian technique involves the merging of a Gaussian filtering technique with the Laplacian technique for edge detection. The technique involves three main steps-filtering, improvement and discovery before finally calculating the tumor area using the proposed algorithm.