Leather Species Identification Based on Surface Morphological Characteristics using Image Analysis Technique

Abstract

Identification and classification of leathers based on the species becomes valuable and necessary due to concerns regarding consumer protection, product counterfeiting or authenticity issues, and dispute settlement in the leather industry and thus helping in trading standards and protecting endangered species. This is carried out mostly by microscopical examination, though, the use of DNA fingerprinting is a theoretical possibility. Identification of leather species through hair pore pattern by microscopical examination requires expertise, training and experience, and due to involvement of human judgment, subjectivity is inevitable. Recent advancements in instrumental techniques aided image analysis offers a good scope for standardizing objective criteria for species identification. In this study, an automatic recognition of leather species based on the surface hair pore pattern using image processing techniques has been investigated. The signature or distinctive feature of leathers from each of the important raw materials dealt within the leather industry were defined from SEM images using image processing technique in terms of number of hair pores, pore density, type and size of the pores (in terms of area, diameter), inter-pore distance and shape of the pore (in terms of circularity, roundness perimeter). Results of the image analysis revealed that all the four common raw materials have distinctive features in terms of specific parameters. Buffalo leather is characterized by the least pore density, largest pore (11110 sq ìm) and highest inter-pore distance (103sq ìm) whereas sheep skin has the smallest hair pore (200 sq ìm) with lowest porosity. Goat skin has both medium (1256 sq ìm) and small sized pores (364 sq ìm) arranged in clusters with trios pattern arrangement and cow leather is characterized by large hair pore density (2262/sq cm) with more uniform mid-sized pores (2190 sq ìm). Thus the developed image processing technique using the current state of art has the potential to provide quantitative estimates of the leather surface morphological characteristics in terms of quantifiable parameters to identify the animal origin.