Inspection apparatus, injection molding system, and inspection method
AssigneeSumitomo Heavy Industries, Ltd.
InventorItta Nozawa
Life of this patent
Estimating the term…
Abstract
An inspection apparatus includes an imaging device acquiring a polarization image of a molding product, and a determination unit determining quality of the molding product. The determination unit includes an image generating unit generating a pseudo image from an input image in accordance with a machine-learned learning model, and the determination unit inputs the polarization image or a calculation image obtained by calculation from the polarization image to the image generating unit as the input image and determines the quality of the molding product based on the input image and the pseudo image generated by the image generating unit.
Claims12
An inspection apparatus comprising:
- an imaging device configured to acquire a polarization image of a molding product, the polarization image having a plurality of channels with different phases;
- and
- a determination unit configured to determine quality of the molding product,
- wherein the determination unit includes an image generating unit configured to generate a pseudo image, which has a plurality of channels, from an input image in accordance with a machine-learned learning model, and
- the determination unit is configured to input the polarization image or a calculation image obtained by calculation from the polarization image to the image generating unit as the input image and determine the quality of the molding product by comparing the input image and the pseudo image generated by the image generating unit in the same channel.
An inspection method comprising:
- acquiring a polarization image of a molding product with an imaging device, the polarization image having a plurality of channels with different phases;
- inputting the polarization image or a calculation image obtained by calculation from the polarization image to an image generating unit as an input image and generating a pseudo image having a plurality of channels from the input image in accordance with a machine-learned learning model;
- and
- determining quality of the molding product by comparing the input image and the pseudo image generated by the image generating unit in the same channel.
Dependencies are inferred from the claim text.