Abstract
AbstractThis paper emphasizes the automation of quality inspection particularly using deep learning algorithm for visual inspection in identifying the surface defects of the impeller. The data collected from one of pump industry was processed using convolutional neural network-based binary classification model. The impeller test images were labeled and augmented and finally inspected with the help of convolutional neural network model. The inspected test images of the impeller have achieved an accuracy of 70.4% which were verified with receiver operating curve.KeywordsDeep learningQuality inspectionMachine vision systemAutomationPythonConfusion matrixROC curve
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have