Defect detection method and device

An electronic device, comprising:

• a processor;
• and
• a memory storing a computer program that, when executed by the processor, causes the processor to:
• obtain a first screen image and a second screen image of a screen module, wherein the first screen image is an image displayed by a display screen collected when the display screen of the screen module is in a first test state, and the second screen image is an image displayed by the display screen collected when the display screen is in a second test state;
• in response to determining a defect area in the first screen image, determine an image feature correlation between the first screen image and the second screen image, including:
• determining an image feature similarity between a first defect image area from the first screen image and a second defect image area from the second screen image, including:
• inputting the first defect image area and the second defect image area into a similarity matching model to obtain the image feature similarity between the first defect image area and the second defect image area, the similarity matching model being trained based on a plurality of pairs of image samples marked with similarity matching results, and the first defect image area including the defect area, wherein:
• the defect area indicates that the screen module has a display defect, and
• the image feature correlation indicating that the image features between the first screen image and the second screen image are similar includes the image feature similarity exceeding a predetermined threshold;
• and
• determine a defect detection result of the screen module based on the image feature correlation.

A defect detection method, comprising:

• obtaining a first screen image and a second screen image of a screen module, wherein the first screen image is an image displayed by a display screen collected when the display screen of the screen module is in a first test state, and the second screen image is an image displayed by the display screen collected when the display screen is in a second test state;
• in response to determining a defect area in the first screen image, determining an image feature correlation between the first screen image and the second screen image, including:
• determining a category feature between a first defect image area from the first screen image and a second defect image area from the second screen image, including:
• inputting the first defect image area and the second defect image area into a category comparison model to obtain the category feature output by the category comparison model, the category feature being used to indicate whether a content of the first defect image area has a same feature with a content of the second defect image area, and the first defect image area including the defect area, wherein the defect area indicates that the screen module has a display defect;
• and
• based on the category feature, determining a target category to which the first defect image area and the second defect image area belong, including:
• inputting the category feature into a classification mode to obtain the target category output by the classification model, the category comparison model and the classification mode being trained based on a plurality of pairs of image samples marked with belonging categories, the target category being one of a first category and a second category, the first category indicating that content features of the first defect image area and the second defect image area are same, and the second category indicating that the content features of the first defect image area and the second defect image area are different, wherein the image feature correlation indicating that the image features of the first screen image and the second screen image are similar includes that the target category is the first category;
• and
• determining a defect detection result of the screen module based on the image feature correlation.