Notice of Violation of IEEE Publication Principles: Sketch Based Image Retrieval System Using ExHoG
G Amuthavalli, G. Prem Sunder, U. Palani, D. Saravanan, Dr.D. StalinDavid, G. Leema Roselin · 2021 International Conference on System, Computation, Automation and Networking (ICSCAN) · 2021
Over the past several years, one of the most vibrant study topics in the field of image processing has been Content-Based Visual Information Retrieval (CBVIR) or Content-Based Image Retrieval (CBIR). The majority of today's picture search engines, such as Google Picture Search and Yahoo Images, rely on image annotations in the form of text. In all of these technologies, photos are often manually labeled with keywords before being retrieved using text-based search approaches. As a result, the performance of these systems is inadequate. The goal of this research project is to design and construct a CBIR system based on free hand sketching. We sought to explain a feasible technique of designing an image retrieval system using a job specific descriptor, which can manage both the informational gap between a free hand drawing and a coloured picture, allowing for an efficient searching mechanism, with the assistance of current approaches. Following a series of pre-processing stages, the descriptor is used to compare the converted full colour picture to the input picture. Many feature extraction approaches are used in practise, such as Histogram of Gradients (HG) and Histogram of Oriented Gradients (HOG), among others. ExHoG (Extended Histogram of Gradients) is being utilized as a new feature for SBIR by investigating the limits of Histogram of Gradients (HG) and Histogram of Oriented Gradients (HOG).