Maltese Christian Statue (MCS) Classification Dataset

Introduction

The `Maltese Christian Statue (MCS) Classifier` project explores the question: Can Artificial Intelligence (AI) be utilised to recognise and differentiate between Maltese Christian statues in images? Built from the curated `MCS Image Classification Dataset`, which represents 17 distinct categories of Maltese Christian statues, this project aims to assist those unfamiliar with the culture or religion by offering an accessible window into Malta's rich religious heritage. `Image classification` is a fundamental task in computer vision, involving the process of categorising images into predefined classes or categories. It leverages machine learning algorithms to analyse the visual content of images and assign them to appropriate labels based on their features and characteristics. In the context of the MCS Classifier project, image classification techniques are employed to automatically identify and categorise Maltese Christian statues depicted in images. This initiative aims to safeguard and promote Maltese religious culture, especially during the solemn period of Lent. It serves as a bridge, introducing tourists to the intricacies of Maltese religious iconography, fostering understanding and appreciation.

Employing advanced image classification techniques, this project integrates artificial intelligence into the context of Maltese Christianity, a domain where such technology has been traditionally less explored. It is essential to underscore that the project is not intended to mock or disrespect religious beliefs. On the contrary, it adopts a respectful and reverent approach, aiming to enrich understanding and foster deeper engagement with Malta's religious heritage.

Ultimately, the project aspires to contribute positively to the perpetuation and enrichment of Maltese religious heritage, potentially inspiring greater belief and dedication to its cause.

MCS Dataset

The `MCS Dataset` features `17` categories of Christian statues found in Malta, specifically in the parish church of `a-ebbu` dedicated to `St Philip of Agira`, and some photos from other parishes. Please note that the images retrieved for the creation of this dataset were extracted from public domain sources and are not intended for commercial use. The categories in the MCS Dataset are: 1. `Christmas Cribs` 2. `Jesus has Risen` 3. `Jesus praying in Gethsemane` 4. `Our Lady of Grace` 5. `Saint Joseph` 6. `Saint Philip of Agira` 7. `Simon of Cyrene` 8. `The Betrayal of Judas` 9. `The Cross` 10. `The Crucifixion` 11. `The Ecce Homo` 12. `The Flogged` 13. `The Lady of Sorrows` 14. `The Last Supper` 15. `The Monument` 16. `The Redeemer` 17. `The Veronica` Sample images from the MCS Dataset are displayed below:

MCS Dataset Distribution

The `MCS Dataset` consists of `5,000` images distributed across the `17` classes. Illustrated below is the distribution of the dataset across the classes. Additionally it is also important to note that the dataset is split into `80%` training and `20%` testing sets. Furthermore, `Data Augmentation` techniques were also used to increase the size of the dataset.

MCS Classifier Model Predictions

Illustrated below are predictions made by the MCS Classifier Model on unseen images from the test dataset. The model demonstrates its ability to classify Maltese Christian statues accurately.

Installation

To get started, clone the repository and navigate to it: bash git clone https://github.com/mbar0075/Maltese-Christian-Statue-Classifier.git cd Maltese-Christian-Statue-Classifier <!-- You can also clone the environment used for this project using the environment.yml file provided in the Requirements directory. To do so, you will need to have Anaconda installed on your machine. If you don't have Anaconda installed, you can download it from here. Once you have Anaconda installed, you can run the following commands to install the environment and activate it

To install the environment, run the following command: bash cd Requirements conda env create -f environment.yml conda activate MCS

Alternatively you can create the environment manually by running the following commands and install the packages in the requirements.txt file in the Requirements directory: bash cd Requirements conda create --name MCS python=3.9.16 conda activate MCS pip install -r requirements.txt

In case you want to install the packages manually, you can do so by running the following commands: