The steel industry faces a significant challenge in classifying steel sheets, which is currently done manually at a considerable cost. To address this issue, we are developing a solution that leverages image analysis techniques to automatically classify steel sheets based on their visual attributes.
#Objectives:- The primary objective of this project is to create an automated system capable of accurately classifying steel sheets using image data. By eliminating the need for manual classification, we aim to reduce costs and improve efficiency in the steel industry.
For image processing:Sklearn,Keras,Numpy,TensorFlow library are used. • Used CNN architecture - ResNet50 and Sequential CNN Keras models with pre trained weights on ImageNet and fine tuned end layers on dataset. • Compared performance of ResNet50 and Sequential CNN Keras model on Precision, Recall, Accuracy, Validation loss.