Development of a machine vision system for leather industry

Leather is one of the oldest protective materials known to man and remains by far the most common upper material for footwear. The greatest variable that confronts both the tanner and the leather product manufacturer, e.g. footwear, is the degree of severity and the distribution of damages caused to the skin of the animals during their lifetime. The subjective assessment of finished leather hide surface quality has resulted in much argument and wasted costs between tanners and footwear industry due to disputed quality standards. This is partly because finished leather hide is large, typically 2.5 x 3.5 m in size, scars on the hide are randomly oriented, and in many instances, healed, only to reappear under tension once the shoe parts are cut and formed. This paper presents the development of a machine vision system for leather industry which enables the computerization of not only the inspection of flaws on a leather hide surface, but also provides the essential data for automatic nesting, cutting and sorting of leather parts, enabling the leather industry to operate in a shared-manufacturing environment. Principal design features for the developed production prototype, software development, and the integration of the developed machine vision system into leather industry infrastructure will be presented.