A machine learning color-based segmentation for object detection within dual X-ray baggage images

Billions of suitcases and other belongings are checked every year in the X-ray systems of airports around the world. This process is of great importance because it involves the detection of possible dangerous objects such as weapons or explosives. However, the work done by airport surveillance personnel is not free from errors usually due to tiredness or distractions. This is a security problem that can always be reduced with the help of automatic intelligent tools. This paper proposes a machine learning (ML) application for image segmentation. First, it is used a color-based pixel segmentation of images to separate organic, inorganic, mixed and opaque objects from the background. Second, those five types of images are reduced in the so-called fusion phase and classified into only two: organic and inorganic. A comparative study of several ML algorithms with heuristics over a large data set of X-ray images is presented for the classification of organic and inorganic objects for a future dangerous object detection work.