This study proposes a novel approach for brain tumor classification by integrating Gray-Level Co-Occurrence Matrix (GLCM) and Local Binary Pattern (LBP) features, along with interaction and composite features. The method enhances the discriminative capability of extracted features through the outer product of GLCM and LBP feature vectors, generating interaction features that capture the interplay between spatial and local texture data. Additional features include aggregated, statistical, and non-linear features derived from GLCM, further enriching the feature space. The proposed approach combines these features with a linear support vector machine (SVM) classifier, achieving a high accuracy of 99.84% in classifying brain tumors into three categories: Glioma, Meningioma, and Pituitary Tumor. The model demonstrates exceptional performance in terms of precision, recall, F1-score, and Dice Similarity Coefficient (DSC), with a DSC of 99.6 and a False Positive Rate (FPR) of 0.00625. The model's performance is validated using the Figshare dataset, showing superior accuracy compared to existing methods. The study highlights the effectiveness of the proposed composite feature extraction model in enhancing the precision of brain tumor classification. The methodology offers a promising solution for improving the accuracy of medical image processing and aiding clinicians in providing more accurate diagnoses and treatments for brain tumors. The research contributes to the field by introducing a novel feature extraction technique that combines GLCM, LBP, and composite features, enhancing the ability to distinguish between different tumor types. The study also emphasizes the importance of feature selection and the integration of advanced machine learning techniques in medical image analysis.This study proposes a novel approach for brain tumor classification by integrating Gray-Level Co-Occurrence Matrix (GLCM) and Local Binary Pattern (LBP) features, along with interaction and composite features. The method enhances the discriminative capability of extracted features through the outer product of GLCM and LBP feature vectors, generating interaction features that capture the interplay between spatial and local texture data. Additional features include aggregated, statistical, and non-linear features derived from GLCM, further enriching the feature space. The proposed approach combines these features with a linear support vector machine (SVM) classifier, achieving a high accuracy of 99.84% in classifying brain tumors into three categories: Glioma, Meningioma, and Pituitary Tumor. The model demonstrates exceptional performance in terms of precision, recall, F1-score, and Dice Similarity Coefficient (DSC), with a DSC of 99.6 and a False Positive Rate (FPR) of 0.00625. The model's performance is validated using the Figshare dataset, showing superior accuracy compared to existing methods. The study highlights the effectiveness of the proposed composite feature extraction model in enhancing the precision of brain tumor classification. The methodology offers a promising solution for improving the accuracy of medical image processing and aiding clinicians in providing more accurate diagnoses and treatments for brain tumors. The research contributes to the field by introducing a novel feature extraction technique that combines GLCM, LBP, and composite features, enhancing the ability to distinguish between different tumor types. The study also emphasizes the importance of feature selection and the integration of advanced machine learning techniques in medical image analysis.