This application use machine learning (convolutional neural network & image processing features) to classify the material corrosiveness. Application link:

1. Download pre-trained model from https://github.com/nurafiqah78/CorrSave/blob/master/pre-trained_model%202019-05-16%2010.37.57
2. skip Tab 1 as you already has the scale attributes ready from the trained model.
3. Download scale attributes file from https://github.com/nurafiqah78/CorrSave/blob/master/Scale%20Attribute.csv
4. Tab 2 (Pre-Train Model): upload (1.) Model File and (2.) ScaleAttribute File
5. Tab 3 (Test Model): Upload your test data (1 image, URL or multiple images in older)

1. Separate your datasets into 2 folders (0: corroded , 1: non corroded)
2. Tab 1 (Model Initialization): set your own parameters
3. Tab 2 (Train Model): upload class 0 and class 1 folders
4. click 'Calculate features' and 'start training'
5. Tab 3 (Test Model): Upload your test data (1 image, URL or multiple images in older)

You may refer to Documentation (https://github.com/nurafiqah78/CorrSave/blob/master/Corrosion_Classification_Documentation.pdf) and shinyapp slide (http:https://rpubs.com/oryzalee/500830) for more info.