Fractal dimension applied in texture feature extraction in x-ray chest image retrieval

摘要

Texture feature extraction is a key step in image retrieval process for nowadays medical images. In this paper, the texture feature extraction of x-ray chest images for image retrieval has been studied. Before texture feature extraction, image preprocessing such as histogram equalization and morphological filtering have been used for improving the contrast and SNR of the image. Meanwhile, some transforms have been done on image matrices to better express the direction details of x-ray chest images. After that, fractal dimension is calculated as the texture feature. The experiment results show that the retrieval system with the fractal algorithm has relatively high precision and recall up to about 85%. Therefore, it can be applied on x-ray chest image retrieval and classification system to help doctors well manage medical images in a large amount.