Inscriptions play a crucial role in preserving historical, cultural, and linguistic information. The identification and analysis of patterns in Tamil letters found in inscriptions provide valuable insights into the evolution of the Tamil language and its script. However, manual analysis of these inscriptions is time-consuming and prone to errors. In recent years, deep learning techniques have shown promising results in pattern recognition tasks, motivating the exploration of various strategies to identify the patterns of Tamil letters on inscriptions.This paper focuses on leveraging deep learning algorithms for the automated identification of Tamil letter patterns in inscriptions. Firstly, a dataset of digitized Tamil inscriptions is collected, consisting of high-resolution images representing a wide range of letter variations. Preprocessing techniques are employed to enhance the quality and clarity of the images, removing noise and artifacts.Various deep learning models, such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), are then trained using the preprocessed image dataset. CNNs excel in extracting spatial features from images, enabling the recognition of letter shapes and contours. RNNs, on the other hand, capture temporal dependencies within sequences of letters, aiding in deciphering the structure and connectivity of the inscriptions. To improve the performance of the models, data augmentation techniques are employed to increase the dataset size and enhance its diversity. However, preprocessing plays a major role in sharpening the features present in the image. Hence, this paper addresses the preprocessing techniques such as Image Blur, Binarization and Edge Detection with respect to inscription. Preprocessing techniques were identified and tested with the inscription image. Based on the results and response time, it is suggested that the Median filter with canny edge detection is working well for inscription images. After preprocessing the results have been tested with edge detection and it is found that, Median filter with canny edge detection gives best accuracy in comparison with other algorithms.