So this project involves developing a machine learning model to detect defective objects using image data. The dataset is split into training, validation, and test sets, with images of both defective and non-defective objects. The images are preprocessed and normalized before being fed into a neural network model built with TensorFlow and Keras. The model includes layers for convolution, pooling, and dense connections, culminating in a binary classifier. The training process involves optimizing the model using the Adam optimizer and binary cross-entropy loss. After training, the model's performance is evaluated on a separate test set. Additionally, the model's effectiveness is demonstrated by predicting the defect status of custom test images, displaying the results alongside the images. The project aims to provide an accurate and efficient method for identifying defects in objects, potentially useful in various industrial applications.
so, here we have trained our model with screw washers so that it will identify defects such as colour, scratch or marking on the screw washer.
ouput:
here we can find some scratches so, that is detected as defected one.
Contributors
- S Sajeev
- A Prithvi
- D Preetham
- Vijay Srinivas
- S Dhanush