Method and apparatus for determining information about a drug-containing vessel
AssigneeAres Trading S.A.
InventorsRobert Jones, Robert F. Marchington, David Rimmer, Bogdan-Alexandru Cociu, William D. Addison, Matthew J. Hayes, Ruairi Duffy, Quentin Le Masne
Life of this patent
Estimating the term…
Abstract
Information about a drug-containing vessel is determined by capturing image data of the curved surface of a cylindrical portion of a drug-containing vessel. The image data is unfurled from around the curved surface, binarised, and a template matching algorithm employed to determine that the label information comprises candidate information about the vessel and/or the drug.
Claims11
A computer-implemented method of determining information about a drug-containing vessel and/or the drug itself, wherein the vessel is located within a medical device and has a cylindrical portion bearing label information about the vessel and/or the drug, the method comprising:
- capturing, using one or more imaging devices contained within the medical device, image data of the curved surface of the cylindrical portion;
- creating a two-dimensional unfurled image from the image data by identifying where points in the unfurled image would map to on the curved surface if the unfurled image was furled about the curved surface;
- binarising the unfurled image;
- applying a template matching algorithm to the binarised image to determine the presence in the binarised image of one or more templates corresponding to candidate information about the vessel and/or the drug;
- and
- based on the determination of the presence in the binarised image of the one or more templates, determining that the label information comprises the candidate information,
- wherein the template matching algorithm is arranged to evaluate each of the templates against the binarised image by evaluating a plurality of tile portions of the respective template against the binarised image, and
- wherein, for each template, a voting image having the same dimensionality as the binarised image is created and its pixels populated based on votes from each of the plurality of tile portions of that template as to whether or not the evaluation of the respective tile portion was above or below a predetermined threshold for that pixel, and the maximum pixel value of each voting image is used to determine the presence in the binarised image of that template, and
- wherein each voting image is blurred prior to determining its maximum pixel value.
A non-transitory computer readable medium comprising machine readable instructions arranged, when executed by one or more processors, to cause the one or more processors to carry out a computer-implemented method of determining information about a drug-containing vessel and/or the drug itself, wherein the vessel is located within a medical device and has a cylindrical portion bearing label information about the vessel and/or the drug, the method comprising:
- capturing, using one or more imaging devices contained within the medical device, image data of the curved surface of the cylindrical portion;
- creating a two-dimensional unfurled image from the image data by identifying where points in the unfurled image would map to on the curved surface if the unfurled image was furled about the curved surface;
- binarising the unfurled image;
- applying a template matching algorithm to the binarised image to determine the presence in the binarised image of one or more templates corresponding to candidate information about the vessel and/or the drug;
- and
- based on the determination of the presence in the binarised image of the one or more templates, determining that the label information comprises the candidate information,
- wherein the template matching algorithm is arranged to evaluate each of the templates against the binarised image by evaluating a plurality of tile portions of the respective template against the binarised image, and
- wherein, for each template, a voting image having the same dimensionality as the binarised image is created and its pixels populated based on votes from each of the plurality of tile portions of that template as to whether or not the evaluation of the respective tile portion was above or below a predetermined threshold for that pixel, and the maximum pixel value of each voting image is used to determine the presence in the binarised image of that template, and
- wherein each voting image is blurred prior to determining its maximum pixel value.
Dependencies are inferred from the claim text.