Content-based image retrieval (CBIR) is a method of developing image search engines. It involves searching for images that are similar to a given query image from a large collection of images. This is done by analyzing the content of each image pixel. CBIR is considered a substitute for traditional text-based image retrieval (TBIR), which depends on text annotation for conducting image retrieval. Although TBIR is capable of conducting image retrieval, it is limited by its use of a restricted vocabulary for annotating images, which may not encompass all aspects of the image’s content. This limitation can lead to ambiguity and various interpretations. This results in the imprecise retrieval of images via TBIR. To address these limitations and overcome additional obstacles, instead of using a textual description, CBIR (Content-Based Image Retrieval) utilizes the content of each image to generate a vector description by analyzing the digital image properties such as color, texture, and shape. This work explores the utilization of specific local texture patterns for image vector description. In addition, we employ various similarity/dissimilarity metrics to evaluate the performance of each technique in comparison to the Corel-1k dataset.