Automatic detection of foreign objects in computed radiography

This paper presents an effective two-step scheme for automatic object detec tion in computed radiography (CR) images. First, various structure elements of the morphological filters, designed by incorporating available morpholo gical features of the objects of interest including their sizes and rough s hape descriptions, are used to effectively distinguish the foreign object c andidates from the complex background structures. Second, since the boundar ies of the objects are the key features in reflecting object characteristic s, active contour models are employed to accurately outline the morphologic al shapes of the suspicious foreign objects to further reduce the rate of f alse alarms. The actual detection scheme is accomplished by jointly using t hese two steps. The proposed methods are tested with a database of 50 hand- wrist computed radiographic images containing various types of foreign obje cts. Our experimental results demonstrate that the combined use of morpholo gical filters and active contour models can provide an effective automatic detection of foreign objects in CR images achieving good sensitivity and sp ecificity, and the accurate descriptions of the object morphological charac teristics. (C) 2000 Society of Photo-Optical Instrumentation Engineers. [S1 083-3668(00)00704-8].