Artificial intelligence techniques for image pattern recognition and retrieval are pivotal in various applications, particularly in medical imaging. Despite numerous existing methods, the process remains complex and computationally intensive. This study introduces an innovative approach to content-based image retrieval (CBIR) by integrating Local Average Binary Patterns (LABP) and the joint probability distribution of color channels. LABP extends the traditional Local Binary Pattern (LBP) by considering multiple layers of neighboring pixels, enabling a more comprehensive texture representation. Additionally, we propose a novel color feature extraction method based on the discrete joint probability distribution of RGB color channels, providing a robust representation of color information. The effectiveness of the proposed method is validated on the Wang (Corel-1k) and Corel-10k datasets, demonstrating superior precision compared to other state-of-the-art techniques. This work contributes to enhancing CBIR performance by combining these novel features into a unified feature vector, improving efficiency and accuracy, especially in large datasets. The code and links to datasets are publicly available at https://github.com/BU-AILab/LABP.