Intelligent image search engine with AI-Based similarity detection for web application
With the growing reliance on visual data and the vast amount of information available on the internet, efficiently searching and retrieving relevant images has become increasingly important. Traditional image search engines which majorly depend on keyword-based approaches often give unsatisfactory r...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Year Project / Dissertation / Thesis |
| Published: |
2024
|
| Subjects: | |
| Online Access: | http://eprints.utar.edu.my/6810/1/2106577_CHONG_WAI_SOON.pdf http://eprints.utar.edu.my/6810/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | With the growing reliance on visual data and the vast amount of information available on the internet, efficiently searching and retrieving relevant images has become increasingly important. Traditional image search engines which majorly depend on keyword-based approaches often give unsatisfactory results since they cannot accurately understand what users mean by their queries in just text. This project aims at developing an intelligent image search engine that uses AI-based similarity detection to have more precise and relevant image retrieval. The project involves developing a web application where users can upload images and search for images in each database that are visually like them. The system uses advanced AI techniques such as Convolutional Neural Networks (CNNs) and Siamese Networks to extract features from input images and compares them with pictures already in the database to pick out and retrieve
the most similar ones. This content-based approach does away with the need for describing images using keywords by users thus improving accuracy and relevancy of search results. This project is a web application that uses different
technologies such as HTML, CSS, JavaScript, Python, PyTorch and MongoDB. The front-end of the application allows for user interaction while the backend handles image processing, similarity detection and database management. To achieve high accuracy in similarity detection, this AI model is continuously refined and improved through iterative development methodology. The project has great impact on various fields like e-commerce sites, digital libraries or
social networks where effective and efficient picture retrieval is highly needed. By creating an excellent image search engine that is user-friendly, it promotes AI powered technology progress in addition to laying a foundation for future research and developments within the field of image search as well as recognition.
|
|---|