MSc Thesis Defense " An End-to-End Hybrid Approach to Automatic and Semi-Automatic Image Annotation" By: Roisul Islam Rumi

Abstract:
This thesis presents a new end-to-end hybrid machine learning (ML) and deep learning (DL) approach for semi-automatic image annotation (SAIA) and automatic image annotation (AIA) in industrial assembly line setups. Image annotation refers
to adding descriptive labels or tags to an image to provide information about the objects and features present in the image. On a high level, the proposed system uses the following steps to annotate images. The first step involves using an ML algorithm, Haar cascade, to split an image into smaller regions of interest (ROI) based on the object of interest, in our case, the connectors of a car’s wire harness (WH). The second step involves using a DL model, specifically a Convolutional Neural Network
(CNN), to classify the ROI images. Our proposed system can do this work both semi-automatically. and automatically. In SAIA, the users can specify which connectors they want to annotate by providing only the class label of the connector(s) and in AIA, the users will only provide an unannotated image for annotation. This work is unique as no other work was set up in the same industrial setting as ours, producing outstanding results surpassing other state-of-the-art image annotation models. The
custom hybrid ML and DL framework, as well as building a unique custom dataset related to the automobile industry assembly line, are significant contributions of this work.

Keywords: Image Annotation, Machine Learning, Deep Learning, Computer Vision, Object Detection, Convolutional Neural Network.