Recent innovations in digital image capturing techniques facilitate the capture of stationary and moving objects. The images can be easily captured via high-end digital cameras, mobile phones and other handheld devices. Most of the time, captured images vary compared to actual objects. The captured images may be contaminated by dark, grey shades and undesirable black spots. There are various reasons for contamination, such as atmospheric conditions, limitations of capturing device and human errors. There are various mechanisms to process the image, which can clean up contaminated image to match with the original one. The image processing applications primarily require detection of accurate noise type which is used as input for image restoration. There are filtering techniques, fractional differential gradient and machine learning techniques to detect and identify the type of noise. These methods primarily rely on image content and spatial domain information of a given image. With the advancements in the technologies, deep learning (DL) is a technology that can be trained to mimic human intelligence to recognize various image patterns, audio files and text for accuracy. A deep learning framework empowers correct processing of multiple images for object identification and quick decision abilities without human interventions. Here Convolution Neural Network (CNN) model has been implemented to detect and identify types of noise in the given image. Over the multiple internal iterations to optimize the results, the identified noise is classified with 99.25% accuracy using the Proposed System Architecture (PSA) compared with AlexNet, Yolo V5, Yolo V3, RCNN and CNN. The proposed model in this study proved to be suitable for the classification of mural images on the basis of every performance parameter. The precision, accuracy, f1-score and recall of the PSA are 98.50%, 99.25%, 98.50% and 98.50%, respectively. This study contributes to the development of mural art recovery.

Historical and ancient murals hold clues to life in the past. This depicts the culture, worship styles, and social life of the community. This fascinates researchers and scholars who study historical bindings to showcase the modern world. Over time, most of these historical monuments have not been preserved in their original state. Most of these are affected due to natural climatic conditions, civil wars, and natural disasters. It remains clueless and imaginary about the size, shape, and color of the distorted historical monuments. This results in limitations for historical studies, archeological research, and geographical surveys. In this paper, we have studied the historical places around Pune city. Identified the locations where the monuments have been distorted and reconstructed recently. The construction age, type, color, shape, and size of the monuments are the major parameters of our study. Based on these criteria, we have captured images of these objects through different angles and camera lens. We have collected and categorized these images into folders with the names of historical places. This image dataset contains both captured and augmented images with distinct angles, scales, and directions. It also includes images captured in the daytime and evening with artificial lighting. This image dataset contains a variety of distinct image patterns that are useful as input to train computer-based supervised learning. The machine learning and deep learning algorithms perform efficiently if the input image dataset is large and distinct. Based on the predictive results generated by the machine learning and deep learning models, it is possible to virtually recreate the original monument. This would add a key value to historical research and studies.

Digital image quality often degrades due to various factors such as noise and blur. Many images are affected by these issues, reducing their clarity and accuracy. This degradation is especially problematic for ancient images, significantly hampers the ability to analyze historical documents and artworks. This paper presents a novel approach to both blur detection and deblur ancient images, enhancing their clarity and readability. This research introduces a technique that combines wavelet transform and convolutional neural networks (CNNs) for effective blur identification and deblurring, specifically aimed at restoring blurred ancient images, regardless of the type of blur degradation. This novel approach demonstrated an average accuracy of 98.3% in blur detection on ancient image datasets. The performance of deblurring algorithms is typically evaluated using metrics such as peak signal-to-noise ratio (PSNR), mean squared error (MSE), and structural similarity index (SSIM) which quantify fidelity and quality of the deblurred images. In the deblurring, this approach produced PSNR values of 55.5 to 68.3 dB, MSE values of 2.99 to 11.1, and an SSIM of 0.9 across different types of blurs. These results show significant promise for the restoration of ancient images, providing researchers, historians, and archaeologists with valuable tool for conservation cultural heritage.