As the American Society of Quality reports, many organizations have quality-related costs of up to 40% of their total production revenue. A large part of this cost comes from inefficiency of manual inspection, which is the most common way to provide quality control in manufacturing. MobiDev AI experts explore the advantages of unsupervised learning for defect detection in manufacturing.
Unsupervised machine learning algorithms allow you to find patterns in a data set without pre-labeled results and discover the underlying structure of the data where it is impossible to train the algorithm the way you normally would. Having a vast experience in machine learning, Mobidev engineers have conducted an experiment using Concrete Crack dataset. The goal was to create a model capable of recognizing images with defects and normal ones using unsupervised learning. Five different approaches were used to get classification results from an unsupervised learning model:
- Clustering
- Birch Clustering
- Custom Convolutional Autoencoder
- DCGAN
- GANomaly
Perhaps, the most beneficial side of unsupervised learning techniques is that we can avoid gathering tremendous amounts of sample data, and labeling it for training. Applying unsupervised learning techniques to derive data patterns, we’re not limited towards which model can be used for actual classification and defect detection.
Read in detail about MobiDev’s approach to Defect Detection with Unsupervised Learning at:https://mobidev.biz/blog/defect-detection-in-manufacturing-with-unsupervised-learning
MobiDev is an international software engineering company with offices in Poland and Ukraine.The company is focused on helping visionaries create their products. MobiDev invests into technology research and has years of experience building AI-powered solutions, implementing machine learning, augmented reality, and IoT.
More Articles
