Compositional Learning of Image-Text Query for Image Retrieval

Image retrieval can be divided into two types context-based and the content-based. Image retrieval based on the content refers to the image features such as color, texture, shape, semantics or sensations. This paper addresses the content-base image retrieval system based on expression sensitivity. It can be image or text query for input the system. Based on Itten theory, expression sensitivity consist of warm, cold, relax, anxious, and life. The research system uses two fuzzy inference system. Firstly, fuzzy inference system is used to decide image region of color. The image size is 256 x 256 pixel. Output the first fuzzy inference system is input for the second fuzzy inference system. The second fuzzy inference system is used to determined expression sensitivity of image. Degree of accuracy based on respondent from 50 images and 20 respondents is 42% for text query and 55% for image query. The further research, it can be used for other image such as medical image with certain criteria.

Download Full-text

Haar Wavelet Untuk Ekstraksi Fitur Energi, Standar Deviasi, Dan Histogram Dalam Sistem Temu Kembali Citra

Jurnal Teknologi Informasi, Komputer, dan Aplikasinya (JTIKA ) ◽

10.29303/jtika.v2i1.67 ◽

2020 ◽

Vol 2 (1) ◽

pp. 40-49 ◽

Cited By ~ 1

Author(s):

Adi Sugita Pandey ◽

I Gede Pasek Suta Wijaya ◽

Fitri Bimantoro

Keyword(s):

Image Retrieval ◽

Image Database ◽

Haar Wavelet ◽

Distance Method ◽

Processing Application ◽

Training Images ◽

Color Features ◽

User Request ◽

Text Query ◽

Database Content

Image retrieval initially uses a query in the form of text to search for images in the database. Image search using text query has a weakness because of the limited description of information stored or given by humans to the metadata on an inconsistent image that greatly affects the duration of searching an image in a database. Content based image retrieval (CBIR) is an image processing application to find the image sought in a large image database based on a query or user request. CBIR technique utilizes features that exist in images, namely color, texture, and shape. These features will be used as a basis for searching images in an image database. In this study the authors used the Haar wavelet method and histogram to look for texture and color features in the image. Then the features found are matched with features stored in the database using the Euclidian distance method. In this study the authors used the Corel dataset as research material. The dataset used is classified into 3 categories: bus, animal and sunset. Each category consists of 100 images where 70% are training images and 30% are test images.

Download Full-text