Accurate diagnosis of the size and location of the tumor plays an important role in identifying the tumor and then using the appropriate treatment course for patients with brain tumors. The innovation of the current research is the use of image processing along with machine learning algorithms to diagnose brain tumors and provide useful information about the tumor so that a suitable treatment course can be drawn for the patient according to this information. In this research, a new structure based on LVQ was presented for the detection of cancerous tumors using MRI images. The proposed method is a hybrid method based on image processing and machine learning algorithms. After applying filters and removing image noises, the images without noise were pre-processed in two categories, bright images and dark images