Smiths Heimann Sas

Other

Smiths Detection France S.a.s.

Najib Gadi

Pascal DeSaute

Laurent Cohen

Irene Dorion

Irène Dorion

Abraham Marciano

Youssef Kacer

Nicolas Dumay

Jean-Olivier Fer

In one embodiment there is provided a method for inspecting a container, comprising: classifying an inspection image of the container in a matching class of one or more predetermined classes of containers of interest, each predetermined class comprising reference images associated with a type of containers of interest, wherein the inspection image is generated using transmission of inspection radiation through the container; comparing a shape model of the inspection image to corresponding shape models associated with reference images within the matching class; associating the inspection image with a matching reference image, based on the comparison; registering one or more zones of the inspection image with corresponding one or more zones of the matching reference image; and mapping differences between the inspection image and the matching reference image, based on the registration.

Detection of irregularities using registration

In one aspect, there is provided a method of denoising one or more inspection images comprising a plurality of pixels, comprising: receiving an inspection image generated by an inspection system configured to inspect one or more containers, the inspection image being corrupted by a Poisson-Gaussian noise and a variance of the noise being non-constant in the plurality of pixels, and denoising the received inspection image by applying, to the inspection image, a variance-stabilizing transformation for transforming the variance of the noise into a constant variance in the plurality of pixels, wherein the variance-stabilizing transformation is based on a descriptor associated with the angular divergence of the inspection radiation and the Poisson-Gaussian noise.

Denoising and/or zooming of inspection images

In one embodiment, the disclosure relates to a method for determining a degree of homogeneity in one or more inspection images of cargo in one or more containers, comprising: determining whether a zone of interest in one or more processed inspection images comprises one or more patterns, wherein the one or more processed inspection images are processed from one or more inspection images generated by an inspection system configured to inspect the one or more containers; and in the event that one or more patterns is determined and that a variation in the determined one or more patterns is identified, classifying the one or more inspection images as having a degree of homogeneity below a predetermined homogeneity threshold.

Determination of a degree of homogeneity in images

In one embodiment, the disclosure relates to a method for inspecting a load () in a container (), comprising: classifying (S) one or more patches () of a digitized inspection image (), the digitized inspection image () being generated by an inspection system () configured to inspect the container () by transmission of inspection radiation () from an inspection radiation source () to an inspection radiation detector () through the container (), wherein the classifying (S) comprises: extracting (S) one or more texture descriptors (V, P) of a patch (), and classifying (S) the patch (), by comparing the one or more extracted texture descriptors (V, P) of the patch () to respective one or more reference texture descriptors (Vr, Wr, Pr) corresponding to respective one or more classes () of reference items (), the one or more reference texture descriptors (Vr, Wr, Pr) of each class of reference items () being extracted from one or more reference images () of the one or more reference items ().

Identification or determination of a load based on texture

Growing community of inventors

Vitry sur Seine, France

Najib Gadi