A circuit board manufacturer wants a vision system that can look at photos of assembled boards and flag three defect types unique to its own product line, such as a specific solder bridge pattern that only appears on its custom connectors. A general-purpose image tagging service returns labels like 'circuit board' and 'metal' but has no concept of these specific defect types. What should the manufacturer do?
Select an answer to reveal the explanation.
Short Explanation
The key detail is that these defects are specific to this one manufacturer's own connectors, so a generic vision service that only knows broad categories like 'metal' or 'circuit board' was never taught what these particular flaws look like. The fix is to feed the system its own examples, photos labeled by defect type, so it can learn categories nobody else's model would recognize. Reading text off the board doesn't help, because a solder bridge isn't a word or a character, it's a shape and a pattern. And a tool built for tracking numeric sensor readings over time has nothing to look at here, since there's no time series involved, just still photographs. When the categories you need are unique to your own product, you train your own model on your own labeled images rather than relying on a general one.
Full Explanation
The correct answer is C. Training a custom image classification model lets the manufacturer supply its own labeled photos of each specific defect type, so the model learns categories that a general-purpose service was never trained to recognize. Option A is incorrect because the prebuilt Computer Vision API only returns broad, generic tags like 'circuit board' or 'metal' and has no way to distinguish the manufacturer's proprietary defect patterns without additional training. Option B is incorrect because optical character recognition extracts printed or handwritten text from an image, but solder bridge defects are visual patterns, not text, so OCR would not detect them. Option D is incorrect because anomaly detection in this context is designed for numeric time-series data, such as sensor readings over time, not for classifying visual defects within a single photograph. Because the defects are specific to this manufacturer's product and involve image content rather than text or numeric sequences, a custom-trained classifier is the right tool.