Aplikasi Aljabar Vektor dalam Sistem Temu Balik Gambar
Kelompok 44 'S43W79N'
NIM
Nama Anggota
13522053
Erdianti Wiga Putri Andini
13522058
Imanuel Sebastian Girsang
13522062
Salsabiila
About The Program
Our program is designed for comparing an image and several images in a dataset with Content-Based Image Retrieval (CBIR) method. CBIR is a technique used to retrieve images from a database. In CBIR, users input a query image, and the system returns images from the database that are similar to the query image. To identify the most similar images, CBIR assesses the content of the query image in comparison to the images stored in the database. Our program employs two types of CBIR, color and texture.
How To Use The Program
Clone this repository.
Change directory to src by running cd src.
Run python index.py on your terminal.
Open a new terminal, then change directory to src by running cd src
Run npm run dev on your terminal.
Follow the local link that appears in your terminal.
Upload your dataset folder by pressing the Upload Your Dataset button for getting dataset from local files or Scrape Web for getting dataset from a website.
Upload the image you want to look for its similarities by pressing the Upload Your Image button.
You can choose between color or texture method by pressing the toggle button.
Press the Search button for starting the comparison process.
Wait for a moment, the results with > 60% similarity will appear below the image. If there are no similar images, the message 'Tidak ada gambar yang mirip!' will appear.
To use camera feature, you can go to Camera page.
Upload your dataset first to activate the camera, then the comparison will running like as before.
You can save the results into a PDF file by pressing Save Your Results? button.
Project Features
Features
Status
CBIR with Color Method
Completed
CBIR with Texture Method
Completed
Capturing Image Using Webcam (Bonus)
Completed
Web Scraping (Bonus)
Completed
Video (Bonus)
Completed
Caching Result (Bonus)
Completed
Saving to PDF File (Bonus)
Completed
Extracting Other Features for Texture Such As Energy, Dissimilarity, and Correlation (Bonus)