Defect detection using synthetic data and machine learning

A method for quality control of contact lens packages, the method comprising:

• generating a data set comprising a plurality of augmented images of a contact lens package having digitally implanted foreign matter disposed therein, wherein the digitally implanted foreign matter disposed in one or more of the plurality of augmented images has characteristics different from other ones of the implanted foreign matter, and wherein the characteristics comprise morphology, size, location, opacity, and quantity of foreign matter;
• training one or more quality control models on at least a training subset of the data set, wherein the one or more quality control models are configured to detect the existence of foreign matter;
• testing the trained one or more quality control models on at least a validation subset of the data set;
• capturing image data of a package of a contact lens;
• analyzing, based on the trained and tested one or more quality control models, the image data;
• and
• causing, based on the analyzing, output of a quality control metric indicative of at least an accept or reject condition of the package.

The system of claim 23 comprising:

• a sensor configured for the capturing image data of a package of a contact lens;
• a computing device in communication with the sensor and configured to receive the image data and to effect one or more of:
• analyzing, based on the one or more quality control models, the image data;
• and
• causing, based on the analyzing, output of a quality control metric indicative of at least an accept or reject condition of the package.

A method for quality control of contact lens packages, the method comprising:

• generating a data set comprising a plurality of augmented images of a contact lens package having digitally implanted defects disposed therein, wherein the digitally implanted defects disposed in one or more of the plurality of augmented images has characteristics different from other ones of the implanted defects;
• training one or more quality control models on at least a training subset of the data set, wherein the one or more quality control models are configured to detect the existence of defects;
• testing the trained one or more quality control models on at least a validation subset of the data set;
• capturing image data of a package of a contact lens;
• analyzing, based on the trained and tested one or more quality control models, the image data;
• and
• causing, based on the analyzing, output of a quality control metric indicative of at least an accept or reject condition of the package.