Towards Black-box Attacks on Deep Learning Apps

Datasets used in Sec.IV-E:

1. https://www.kaggle.com/sriramr/fruits-fresh-and-rotten-for-classification

2. https://image-net.org/challenges/LSVRC/2012/index

3. https://www.kaggle.com/virenbr11/pothole-and-plain-rode-images

4. https://www.tensorflow.org/datasets/catalog/tf_flowers

5. https://www.kaggle.com/lantian773030/pokemonclassification

6. https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/DBW86T

7. https://drive.google.com/file/d/1h7ixQbiWj3T8HuA_gksHX1XvTG8ser_x/view?usp=sharing

8. https://image-net.org/challenges/LSVRC/2012/index

9. https://drive.google.com/file/d/1x4DglqpbWHEhmBHwPFW1ts8M9JgQ8ETt/view?usp=sharing

10. https://image-net.org/challenges/LSVRC/2012/index

Results for apps in RQ1:

Results for apps in RQ2:

Results for apps in RQ3:

Pipeline Codes:

https://drive.google.com/file/d/1oYh0gpRdj3A8ehMVIlZQV58EqF2OT4IP/view?usp=sharing