An X-Ray Image Enhancement and Object Detection Using YOLOv8 in Airport Security Inspection

Abstract

Enhancing X-ray images for airport security purposes is introduced in this study. By integrating techniques such as Unsharp Masking (USM) and Contrast Limited Adaptive Histogram Equalization (CLAHE) with YOLOv8 for object detection, the method addresses the issue of color distortion often encountered in CLAHE-enhanced X-ray images. Initially, the grayscale images of the X-ray, broken down into its red, green, and blue channels, undergo CLAHE enhancement individually. Subsequently, the enhanced gray scale images are merged, followed by the application of USM sharpening to further enhance fine details. The resulting sharpened image is then blended with both the original and the USM-sharpened versions, with the weighting determined by specific parameters. Finally, YOLOv8 is employed for object detection, yielding promising results in improving image quality while mitigating color distortion. The USM+CLAHE approach significantly enhances the quality of security X-ray images while effectively suppressing color distortion. By systematically combining advanced image enhancement techniques with state-of-the-art object detection methods like YOLOv8, this study provides a comprehensive solution to the challenges posed by color distortion in CLAHE-enhanced X-ray images. The proposed method not only improves the clarity and detail of X-ray images but also enhances the accuracy and efficiency of object detection, there by contributing to enhanced security measures in airport environments.