Automatically grade the failure modes of the test result images with grading assistance. Optical inspection calculates the area of bulk material remaining as a percentage of the whole. The result and a picture with graphical overlay are stored in the system together with the measurement.
You do not need operator intervention to assist in grading the failure modes of test results. Using deep learning and a self learning library, we train a neural network in image processing. The system learns features which allow the classifier to distinguish between failures. By classifying the failure mode criterium beforehand, image recognition is ready to perform fully automatic grading without any assistance.
Operators do not need to fulfill assessments by accepting or editing the failure modes at the end of a fully automatic run. Deep learning automatically determines the type of pull or shear code based on given failures modes standards, for example, the JEDEC Standard 22-B116. Afterward, it automatically calculates the percentage of the remaining bond material in the region of interest and identifies the failure mode using classifications.
In exceptional cases, a slight intervention of an operator is necessary to classify the grading. Classification errors such as empty, very small, or too large pad(lift)s, gold areas, or bond areas, automatically show notifications for operator identification. You can easily classify and adjust these error messages on a remote workstation during or after the grading.
Dive straight into the feedback!Login below and you can start commenting using your own user instantly