• • • Analyze the how the gaussian noise levels to the dataset impact the model accuracy.

    • Adding the noise from 0 to 1.0 to dataset and observe the model accuracy for each level.

    • Got PSNR and accuracy for the respective noise and analyzed the noise range to get high accuracy using the accuracy vs PSNR graph

• Gausian noise values and dataset samples

0.15 0.20 0.35 0.70

• Accuracy Vs Noise graph

• Up to point 0.7 the gaussian noise much not affect the model accuracy

• Next week target : Run the model with gray scale images to confirm that the main discriminant features were the shapes of the tin cans and tires