A novel generalization of the gray-scale histogram and its application to the automated visual measurement and inspection of wooden Pallets

Abstract

We present a novel concept, the histogram of connected elements (HCE) which is a generalization of the usual gray-level histogram of digital images is introduced and its application to automatic visual measurement and inspection system, currently operating at several European inspection plants. The main objective of the system is to automate the inspection of used wooden pallets. The paper begins with a brief description of the system, including some comments on the electromechanical handling of the inspected objects and the illumination set-up. Then, the paper presents the segmentation method used to extract the pallet elements, as an initial step for pallet measurements and the detection of possible defects. This method consists of an initial threshold on the histogram based on a Bayesian statistical classifier, followed by an iterative, heuristic search of the optimum threshold of the histogram. Finally, the paper introduces the application of the histogram of connected elements to the detection of very thin cracks, one of the hardest problems involved in the visual inspection of used pallets. Experimental results are obtained and we present a comparative study with several well-known and thoroughly tested techniques for the segmentation of textured images, including two algorithms belonging to the adaptive Bayesian family of restoration and segmentation methods and a probabilistic relaxation process.