Content based Image Retrieval System using Hybrid model of optimization for medical databases
Description
This repository showcases the research work conducted to develop an advanced Content-Based Image Retrieval (CBIR) system utilising a hybrid model of optimization specifically designed for medical databases. The repository encompasses various components, including image pre-processing techniques, feature engineering methodologies, and machine learning algorithms, each accompanied by their respective outcomes. The repository is structured into distinct folders, each dedicated to a specific task. The Pre-processing folder houses code implementations for essential image enhancement techniques, such as contrast enhancement, etc aimed at improving the quality and usability of medical images within the CBIR system. The Feature Engineering folder contains code and methodologies (optimisation approaches such as cuckoo) utilised to extract and transform relevant features from the pre-processed medical images. These features encompass diverse characteristics, including colour, etc which are crucial for effective image retrieval and analysis. The Machine Learning folder encompasses the implementation of various machine learning algorithms specifically tailored for medical image analysis and retrieval tasks. These algorithms are employed to train models capable of recognizing and categorising medical images based on their extracted features, enabling accurate retrieval and classification of relevant images. Additionally, the repository includes meta-information detailing the datasets utilised for training and evaluation purposes. This information provides insights into the composition, size, and annotation of the medical image datasets, ensuring transparency and reproducibility of the research work. Overall, this repository serves as a comprehensive resource for researchers and practitioners in the field of medical image retrieval, offering a hybrid model of optimization that combines pre-processing techniques, feature engineering, and machine learning algorithms to enhance the retrieval and analysis of medical images in a content-based manner.