This paper introduces a novel approach for accurately recognizing Indian Sign Language (ISL) gestures, crucial for enhancing communication accessibility for individuals with hearing impairments. Leveraging advanced machine learning techniques, we propose a methodology that combines segmentation, feature extraction, and classification using Local Binary Patterns (LBP) and Gray-Level Co-occurrence Matrix (GLCM). Experimental results demonstrate significant advancements, with our Random Forest classifier achieving an accuracy of 99.45% with LBP features and 95.64% with GLCM features, surpassing existing studies. This research contributes to ISL recognition technology, offering a promising solution to improve inclusivity for the deaf and hard-of-hearing community, with potential for real-world applications.