Skip to content

Latest commit

 

History

History
20 lines (16 loc) · 942 Bytes

README.md

File metadata and controls

20 lines (16 loc) · 942 Bytes

Venice Boat Classifier

Dataset

MarDCT Maritime Detection, Classification and Tracking data set Can be downladed from: http://www.dis.uniroma1.it/~labrococo/MAR/classification.htm

Description

Image classification problem with 21 boat categories, with provided train and test set. Several approaches were evaluated, mainly based on Convolutional neural networks (CNN), with Tensorflow, Keras, and Scikit-learn. To reproduce results, the dataset must be downloaded in the same folder, and follow the steps depicted on HW2-PrepareData.ipynb. Subsequently, HW2-ProcessData.ipynb can be processed.

Resources

https://www.learnopencv.com/keras-tutorial-fine-tuning-using-pre-trained-models/ https://gogul09.github.io/software/image-classification-python https://www.tensorflow.org/tutorials/keras/basic_classification https://keras.io/