Item description:

 Currently, the customer has a daily laptop output of more than 1,500 in a single production line, but their detection speed and accuracy can't meet the standards, and some defects are greatly missed. The customer introduced the traditional visual detection equipment a few years ago, but it still fails to meet the detection standards after years of debugging, with over 20% of miss and error rate. The factory has 100 production lines. With the help of AOI detection equipment, each production line still needs to be staffed with 2 workers for second detection. Due to the large product area, random defect location, complex defect feature extraction, and diverse product types, it is difficult to extract the correct defect features only by AI algorithms. DSTEK combines deep learning and traditional algorithms and enables the latter to assist AI to extract defect features, winning satisfactory results.

 Defect types:

 Screws missed/misaligned/not locked to the end; accessories such as memory/network card/antenna cable/speaker/and other cables or plugs aren't correctly connected.

                                                                                                               Original  picture                                                                                                                                     Detection picture

      AI algorithms are 50 to 80 times faster than the traditional algorithms       Image processing technologyOverall online detection     Overall online detection