**5. Conclusions**

This study presents an automatic brain tumor diagnostic approach using brain MRI images. First, the proposed approach computes the SURF and KAZE features using a grid of 8 × 8 pixels in size of brain MRI images. Then, 80% of the strongest features are considered for segmentation using *k*-means clustering. The final feature vector has a size of 400 per image for each feature (SURF and KAZE). Finally, the proposed hybrid feature vector is used to train the SVM model. The classification accuracies of the proposed model (SURF + KAZE) are 95.33% and 95.9%, almost 2% higher than the SURF-trained SVM model. The comparison of the proposed approach with the findings presented in the literature also shows its superiority due to its high accuracy and lower computational time. Thus, the proposed approach can be used for the automatic detection of brain tumors.

**Author Contributions:** Conceptualization, Y.E.A., M.U.A., W.A. and A.Z.; data curation, S.K.A.; formal analysis, Y.E.A., M.U.A., M.I. and H.A.A.; investigation, A.Z. and S.K.A.; methodology, Y.E.A., M.U.A. and W.A.; project administration, A.Z., S.K.A. and M.A.A.B.; resources, M.I.; software, K.D.K. and A.K.A.; supervision, H.A.A.; validation, M.U.A.; writing—original draft, Y.E.A., M.U.A. and A.Z.; writing—review and editing, K.D.K., A.Z., M.I., M.A.A.B. and A.K.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors are thankful to the Deanship of Scientific Research, Najran University, Kingdom of Saudi Arabia.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors world like to express their gratitude to the Deanship of Scientific Research, Najran University, Kingdom of Saudi Arabia, for their financial and technical support under code number (NU/-/MRC/10/388).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

