Improvement of Security Systems by Enhancing the Detection of Objects in Baggage X-Ray Radiography Images

Abstract

Visual inspection of baggage in airports and other critical centers takes a great time. In many cases, especially when the objects inside the load are positioned at specific angles, they may not be detectable by rapid visual examinations. Nowadays, X-ray imaging devices are used to detect objects at the passengers' luggage. Radiographic images obtained due to scattering of photon have a degree of fogging, and sometimes the exact diagnosis of objects is difficult. Image processing methods can improve the contrast and the ability to recognize objects. Different noises and scattering levels exist in radiographic images, which justify the use of automated image processing methods. In this study, Gabor filter and wavelet methods with automatic threshold level have been used to improve the quality of images. According to the results, there is no significant difference in the degree of detection of the objects, regardless of the differences in the reconstructed images. In both methods, the contrast and detection capabilities of the primary radiography images have improved significantly. The computation time of the implementation of the discrete wavelet algorithm is about one eighth of Gabor's time of execution, which can be remarkable given the importance of image processing speed in the area of passenger traffic inspection.

Keywords

• luggage radiography
• luuggage system inspection
• x-ray
• discrete wavelet method
• gabor filter