Partial Matching Using Set Exclusion Criteria: Applied to Livestock Brand Retrieval

This article presents a partial matching approach based on set exclusion criteria applied to livestock brand retrieval. A set exclusion measure function, depending on local shape features, is formulated. This exclusion measure helps to determine if a registered brand is a subset of another existing brand, or vice-versa. Local features are obtained taking account the relative spatial distribution of brand components. Experimental results have shown the accuracy of the methodology in cattle brand identification.