Comparison of Histogram Features and Co-occurrence Matrix in Identification of Lung Cancer X-Ray Images with Naive-Bayes Method

M. A. Bustomi*, A. M. Rusnandar

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

Abstract

Many studies have been carried out on the identification of objects in the image. One feature often used in object identification is the histogram feature of digital images. The next development is the use of a digital image co-occurrence matrix as an identifier. This study aims to compare the use of two object identifiers in images that are not derived from visible light but are images derived from X-rays. In this study, the X-ray image is an X-ray image of lung cancer, and the object identification method used is the Naive Bayes method. The number of lung X-ray images used was 100 images. The master image is divided into 60 images for training and 40 images for testing. This study's results indicate that using a digital image co-occurrence matrix as an identifier gives better results than the histogram feature. This study concludes that the comparison of histogram features and co-occurrence matrix in X-ray images shows the same results as in visible light images.

Fingerprint

Dive into the research topics of 'Comparison of Histogram Features and Co-occurrence Matrix in Identification of Lung Cancer X-Ray Images with Naive-Bayes Method'. Together they form a unique fingerprint.

Cite this