Ancient stone inscriptions are invaluable cultural artifacts, often bearing historical texts or inscriptions in languages that have long ceased to be in common use. In this study, we propose a novel approach to the analysis and recognition of characters in the stone inscriptions from ancient times using the Histogram of Oriented Gradients technique. In this context, histogram of oriented gradients offers a promising solution for character recognition by capturing the subtle visual cues that define the shapes and textures of individual characters. We present a comprehensive methodology for applying histogram of oriented gradients to ancient stone inscriptions, including image preprocessing, character segmentation, and the calculation of histogram of oriented gradients descriptors for each segmented character. To enhance the recognition accuracy, we also explore the utilization of machine learning classifiers, such as support vector machines trained on histogram of oriented gradients descriptors from labeled datasets of ancient stone inscription characters. Here, digital images of stone inscriptions from the eleventh century are used to create a database. Support Vector Machine is used to classify the features into modern characters after they have been computed. Results show a significant recognition rate for 10 characters out of 300 characters in a database of 300 characters.