Visual search using deep learning

• Main Author: Teo, Peggy
• Other Authors: Yap Kim Hui
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/74773
• Institution: Nanyang Technological University

Object recognition has become an extremely popular topic in today’s world as it is one of the core problems in computer vision. The technology uses an input image instead of words to help people to search for information. It often involves image processing and pattern recognition techniques in recent visual search engines. Image processing and machine recognition of pattern are the ability of a software to identify places, people, animals, objects, handwriting and even moving actions in images. Computers are another platform where it can be used with the addition of a camera and artificial intelligence software. With the rapid progression of technologies, visual search has become a very popular topic in the modern society. Till date, many areas have incorporated the use of image recognition ability, such as security [1], healthcare [2], e-commerce to websites [3], applications, gaming [4] and social networking platforms [5]. Hence, this project is to study the use of deep learning and Convolutional Neural Networks on image recognition. The data set used will be different dog breeds as the image retrieval is convenient and widely available on the internet. However, it can be challenging as the differences among breeds can be subtle. Moreover, the project aims to serve an educational purpose where people can learn about the different dog breeds they come across, help to identify lost dogs, aid organisations such as Society for the Prevention of Cruelty to Animals (SPCA) and other animal shelters recognise and categorise the different dogs that were brought in.