The invention relates to a segmentation method based on the characteristics for segmenting a plurality of duplicate articles (3) arranged loosely, which comprises stages of: acquiring an image (M) of a sample article (30); calculating keypoint-descriptors of the image (M); defining an identifying figure (Z) on the image (M); acquiring a first image (11) of a plurality of duplicate articles; performing a matching of the thus-defined keypoint-descriptor pairs; acquiring a position and an orientation of the identifying figure (Z) with respect to a first keypoint-descriptor pair of the image (M) having a match with a second keypoint-descriptor pair of the first image (11); defining, in the first image (11), an identifying figure of projection as a Euclidean transformation of the identifying figure (Z), with reference to the first and second pairs; applying the two preceding stages to a plurality of keypoint-descriptor pairs of the image (M) having a match with a keypoint-descriptor pair of the first image (11); collecting together identifying figures of projection having between them a predetermined degree of superposing; defining a representative figure for each group of identifying figures of projection which is formed by a minimum predetermined number of identifying figures of projection, which representative figure has a same shape and dimension as an identifying figure of projection, and is selected in order to estimate a position of a corresponding article illustrated in the first image of a plurality of duplicate articles. The invention also relates to a method for picking up articles (3) arranged loosely in a storage zone of articles (5) and for positioning the articles (3) in an outlet station (SU), and a group for actuating the method.
Feature-based segmentation method, for segmenting a plurality of loosely-arranged duplicate articles and a group for actuating the method for supplying a packaging machine / G., Monti; Prati, Andrea; Piccinini, Paolo; Cucchiara, Rita. - .
Feature-based segmentation method, for segmenting a plurality of loosely-arranged duplicate articles and a group for actuating the method for supplying a packaging machine
PRATI, Andrea;PICCININI, PAOLO;CUCCHIARA, Rita
Abstract
The invention relates to a segmentation method based on the characteristics for segmenting a plurality of duplicate articles (3) arranged loosely, which comprises stages of: acquiring an image (M) of a sample article (30); calculating keypoint-descriptors of the image (M); defining an identifying figure (Z) on the image (M); acquiring a first image (11) of a plurality of duplicate articles; performing a matching of the thus-defined keypoint-descriptor pairs; acquiring a position and an orientation of the identifying figure (Z) with respect to a first keypoint-descriptor pair of the image (M) having a match with a second keypoint-descriptor pair of the first image (11); defining, in the first image (11), an identifying figure of projection as a Euclidean transformation of the identifying figure (Z), with reference to the first and second pairs; applying the two preceding stages to a plurality of keypoint-descriptor pairs of the image (M) having a match with a keypoint-descriptor pair of the first image (11); collecting together identifying figures of projection having between them a predetermined degree of superposing; defining a representative figure for each group of identifying figures of projection which is formed by a minimum predetermined number of identifying figures of projection, which representative figure has a same shape and dimension as an identifying figure of projection, and is selected in order to estimate a position of a corresponding article illustrated in the first image of a plurality of duplicate articles. The invention also relates to a method for picking up articles (3) arranged loosely in a storage zone of articles (5) and for positioning the articles (3) in an outlet station (SU), and a group for actuating the method.Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris