With the advancements of Deep Learning technologies, its application has broadened into the fields of food classification from image recognition using Convolutional Neural Network, since food ingredient classification is a very important aspect for eating habit recognition and also reducing food waste. This research is an addition to the previous research with a clear illustration for deep learning approaches and how to maximize the classification accuracy to get a more profound framework for food ingredient classification. A fine-tuned model based on the Xception Convolutional Neural Network model trained with transfer learning has been proposed with a promising accuracy of 95.20% which indicates a greater scope of accurately classifying food objects with Xception deep learning model. Higher rate of accuracy opens the door of further research of identifying various new types of food objects through a robust approach. The main contribution in the research is better fine-tuning features of food classification. The dataset used in this research is the Food-101 Dataset containing 101 classes of food object images in the dataset.

Underwater Object Detection is one of the most challenging and unexplored domains in this area of Computer Vision. The proposed research refines the image enhancement of under-water imagery by proposing an improvement of already existing tools for underwater Object detection. The comparative study clearly depicts the enhancement of the proposed method with respect to the existing methods for underwater object detection. Moreover, a framework for detection of underwater organisms such as fishes are proposed, which will act as the steppingstone for re- serving the ecosystem of the whole fish community. Mostly the object detection using deep learning has been the prime goal to this research and the comparison between other preexisting methods are compared at the end. As a result, techniques that are already well established will be used for overall enhancement of those images. Through this enhancement and through finding a healthy environment for their breeding ground, the extinction of selected species of fishes is can be diminished and decreased. All this is carried out by overcoming difficulties underwater through a novel technique that can be integrated into an Underwater Autonomous Vehicle and can be classified as robust in nature. Robustness will depend on three important factors in this research, first is accuracy, then fast and lastly being upgradeable. The proposed method is a modified VGGNet-16, which is trained using the ImageCLEF FISH_TS dataset. The overall result provides an accuracy of 96.4% which surpasses all its predecessors.