Morphology-Based Approach to Detection of Free Form Line Objects in Grayscale Images

  • I. Mednieks Institute of Electronics and Computer Science


A method for detection of free form line objects in grayscale images is proposed, based on several sequential processing steps, partly using operations of mathematical morphology. The steps include background compensation, thresholding to obtain binary images, gluing breaches in binary images of objects, skeletonisation, calculation of object length, and making decision about the presence of the object on the basis of its length. The method can be used e.g. in X-ray systems for detection of contaminants in food packages if the images of contaminant objects take the form of a line object with any shape, e.g. fishbones, small metal wires etc. The paper describes application task, ideas behind the method, processing steps and results obtained by processing real X-ray images obtained by food control systems.