Implementasi Temu Kembali Citra Menggunakan Fitur Warna Berbasis Histogram dan Fitur Tekstur Berbasis Blok

Rahayu, Sani Puji (2017) Implementasi Temu Kembali Citra Menggunakan Fitur Warna Berbasis Histogram dan Fitur Tekstur Berbasis Blok. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
5113100153-Undergraduate_Theses.pdf
Restricted to Repository staff only

Download (4MB) | Request a copy

Abstract

Citra digital biasa digunakan masyarakat dalam berbagai bidang seperti kesehatan, perdagangan, dan hiburan. Hal ini menyebabkan meningkatnya citra digital yang dihasilkan setiap harinya. Citra digital yang dihasilkan kemudian disimpan dalam suatu tempat penyimpanan seperti database. Banyaknya citra digital yang disimpan dalam database menyebabkan sulitnya pengelolaan file-file citra terutama dalam menemukan konten citra yang diinginkan. Content based image retrieval (CBIR) merupakan sebuah metode pencarian citra dengan melakukan perbandingan antara citra query dengan citra yang ada di database berdasarkan informasi yang ada pada citra tersebut. Pada tugas akhir ini, dibangun suatu sistem temu kembali citra menggunakan fitur warna berbasis histogram dan fitur tekstur berbasis blok. Metode histogram warna digunakan untuk ekstraksi fitur warna dan ekstraksi Block Difference of Inverse Probabilities dan Block Variation of Local Correlation Coefficients digunakan untuk mengekstraksi fitur tekstur. Metode Square Chord Distance digunakan untuk menghitung jarak citra. Hasil pencarian citra mirip dengan rata-rata precision terbaik didapatkan dari perpaduan ekstraksi fitur warna dan tekstur warna dengan rata-rata precision 93.71% dan rata-rata waktu komputasi 0.2281 detik. Sedangkan untuk perpaduan ekstraksi dengan hasil rata-rata waktu komputasi terbaik adalah menggunakan perpaduan ekstraksi fitur warna dan tekstur brightness dengan rata-rata precision 92.22% dan rata-rata waktu komputasi 0.1468 detik. ================================================================================================= Digital imagery is commonly used by people in the fields of health, commerce, and entertainment. Digital images are usually stored in a storage place such as a database. The large number of digital images stored in the database causes the difficulty of managing image files, especially in finding the desired image content. Content based image retrieval (CBIR) is an image search method by performing a comparison between the image of the query and the image in the database based on the information contained in the image. In this final project, an image retrieval system were built using histogram-based color features and block-based texture features. Color histogram method is used for color feature extraction. Block Difference of Inverse Probabilities and Block Variations of Local Correlation Coefficients are used to extract texture features. Square Chord Distance method is used to calculate the distance of the image. The best precision of image retrieval were obtained from the combination of color extraction and color texture extraction with average precision 93.71% and the average computation time 0.2281 seconds. As for the best average computation time of image retrieval were obtained from combination of color feature and brightness texture with average precision 93.71% and the average computation time 0.1468 seconds.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Image Retrieval, Color Histogram, Block Difference of Inverse Probabilities, Block Variation Of Local Correlation Coefficients
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
Z Bibliography. Library Science. Information Resources > ZA Information resources > Z699.5 Information storage and retrieval systems
Divisions: Faculty of Information Technology > Informatics Engineering > (S1) Undergraduate Theses
Depositing User: Sani Puji Rahayu
Date Deposited: 09 Nov 2017 07:54
Last Modified: 21 Nov 2017 08:25
URI: http://repository.its.ac.id/id/eprint/42446

Actions (login required)

View Item View Item