This project uses the CIFAR-10, a publicly available dataset containing 60,000 color images across 10 classes. Two images were selected from the dataset: Image 1: Airplane Image 2: Automobile
Size: 32 × 32 pixels Channels: 3 (RGB) Pixel Value Range: [0, 255] The dataset is loaded programmatically using PyTorch’s torchvision library; no manual image upload is required.
The following preprocessing steps were applied to both images: Resize – Images resized to 64 × 64 to standardize input size Grayscale Conversion – Reduced color channels to simplify features Rotation – Applied random rotation to simulate orientation variance Horizontal Flip – Introduced data augmentation Normalization – Scaled pixel values to the range [-1, 1] for AI models Bonus: Custom Sharpening Filter – Enhanced edges using a kernel-based filter
For every preprocessing step: The transformation was applied to both images Outputs were visualized side-by-side
Image shape and pixel value range were printed
Normalization and sharpening were retained for model input, while de-normalization was used only for visualization to ensure clarity.