.png){kind=icon}
.png){kind=icon}
Return
Reducible tissue and principal component fusion based x-ray image boosting for healthcare applications
.png){kind=icon}
PRE
AI
S
A
M
DOI：10.1007/s11042-026-21847-w.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
X-ray imaging is a significant tool to acquire X-ray images for the detection of diseases inside the human body in the medical field. X-ray images are captured from electromagnetic waves, i.e., X-rays. It has low contrast and hazing due to the small amount of EM wave, so the detection of organs and nodules is too hard. Therefore, the dynamic range adjustment for improving the contrast of the X-ray image is a necessary step to detect diseases. A novel method based on component reduction with a fusion technique has been developed for adjusting the dynamic range of the X-ray images. In this technique, an X-ray image is divided into two parts, i.e., tissue and detailed elements. The tissue element is a reducible element that varies according to the reduction coefficient to remove the partial components from an X-ray image. After that, to preserve the brightness consistency, an adaptive parameter is used. The improved image is generated using adaptive parameters and removable components for each reduction coefficient. In the end, a principal component analysis (PCA) fusion technique is applied to all created images using different reduction coefficients. The proposed approach is employed over the X-ray image dataset provided by the Japanese Society of Radiological Technology. The subjective and objective analysis of the proposed approach is calculated to explain the visual quality of the images and the best result acquired to analyze the dynamic range adjustment of the images, respectively. The basic details and edge information are preserved in the enhanced images.
Keywords:
Reducible components
Principal component analysis
Local maximum component
Local minimum component
Dynamic range adjustment
X-ray image
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W
Organization
Cited Papers