Methods and systems for improved quality inspection

A server system, comprising:

• one or more processors;
• and
• memory storing one or more programs for execution by the one or more processors, the one or more programs including instructions for:
• obtaining a plurality of test images, each test image capturing a respective one of a plurality of composite objects comprising two or more components;
• for each of the plurality of test images:
• performing independent object inspection on each of two or more sub-portions of the test image, wherein each sub-portion of the test image corresponds to a respective one of the two or more components of the respective composite object captured in the test image, wherein performing the independent object inspection includes:
• segmenting the test image into at least a first sub-portion of the test image containing a first component and not containing a second component of the two or more components, and a second sub-portion of the test image containing the second component and not containing the first component of the two or more components, including:
• detecting user inputs respectively directed to a first location in the test image and a second location in the test image;
• creating a first mask for the test image to extract the first sub-portion of the test image in accordance with the first location;
• creating a second mask for the test image to extract the second sub-portion of the test image in accordance with the second location;
• and
• extracting the first sub-portion of the test image and the second sub-portion of the test image using the first mask and the second mask, respectively;
• performing object inspection on the first sub-portion of the test image using a first subset of information channels of the test image and a first model trained on a first set of training images containing the first component and not containing the second component;
• and
• performing object inspection on the second sub-portion of the test image using a second subset of information channels of the test image that are distinct from the first subset of information channels, and a second model trained on a second set of training images containing the second component and not containing the first component.

A non-transitory computer-readable storage medium, storing one or more programs configured for execution by one or more processors of a server system, the one or more programs including instructions for:

• obtaining a plurality of test images, each test image capturing a respective one of a plurality of composite objects comprising two or more components;
• for each of the plurality of test images:
• performing independent object inspection on each of two or more sub-portions of the test image, wherein each sub-portion of the test image corresponds to a respective one of the two or more components of the respective composite object captured in the test image, wherein performing the independent object inspection includes:
• segmenting the test image into at least a first sub-portion of the test image containing a first component and not containing a second component of the two or more components, and a second sub-portion of the test image containing the second component and not containing the first component of the two or more components, including:
• detecting user inputs respectively directed to a first location in the test image and a second location in the test image;
• creating a first mask for the test image to extract the first sub-portion of the test image in accordance with the first location;
• creating a second mask for the test image to extract the second sub-portion of the test image in accordance with the second location;
• and
• extracting the first sub-portion of the test image and the second sub-portion of the test image using the first mask and the second mask, respectively;
• performing object inspection on the first sub-portion of the test image using a first subset of information channels of the test image and a first model trained on a first set of training images containing the first component and not containing the second component;
• and
• performing object inspection on the second sub-portion of the test image using a second subset of information channels of the test image that are distinct from the first subset of information channels, and a second model trained on a second set of training images containing the second component and not containing the first component.