app.py

import gradio as gr
import kornia as K
from kornia.core import Tensor
from kornia import morphology as morph
import torch
import numpy as np

def morphological_operators(img, operator, kernel, kernel_size): 
    # Convert input to tensor
    if isinstance(img, np.ndarray):
        img = torch.from_numpy(img).permute(2, 0, 1).float() / 255.0
    elif isinstance(img, torch.Tensor):
        img = img.permute(2, 0, 1).float()
        if img.max() > 1.0:
            img = img / 255.0
    else:
        raise ValueError(f"Unsupported image type: {type(img)}")

    img = img.unsqueeze(0)  # Add batch dimension
    device = 'cpu'  # 'cuda:0' for GPU
    img = img.to(device)

    kernels = {
        "Ones": torch.ones(kernel_size, kernel_size).to(device),
        "Eye": torch.eye(kernel_size).to(device),
        "Cross": torch.tensor([[0, 1, 0],[1, 1, 1],[0, 1, 0]]).to(device),
    }
    operations = {
        'Dilation': morph.dilation(img, kernels[kernel]),
        'Erosion': morph.erosion(img, kernels[kernel]),
        'Open': morph.opening(img, kernels[kernel]),
        'Close': morph.closing(img, kernels[kernel]),
        'Gradient': 1. - morph.gradient(img, kernels[kernel]),
        'Bottom Hat': 1. - morph.bottom_hat(img, kernels[kernel]),
        'Top Hat': 1. - morph.top_hat(img, kernels[kernel])
    }
        
    output = K.tensor_to_image(operations[operator].squeeze(0))
    return output

examples = [
    ["examples/cat.png", "Dilation", "Ones", 3],
    ["examples/huggingface.jpg", "Close", "Eye", 5]
]
title = "Kornia Morphological Operators"
description = "<p style='text-align: center'>This is a Gradio demo for Kornia's Morphological Operators.</p><p style='text-align: center'>To use it, simply upload your image, or click one of the examples to load them, and select any morphological operator to run it! Read more at the links at the bottom.</p>"
article = "<p style='text-align: center'><a href='https://kornia.readthedocs.io/en/latest/' target='_blank'>Kornia Docs</a> | <a href='https://github.com/kornia/kornia' target='_blank'>Kornia Github Repo</a> | <a href='https://kornia.github.io/tutorials/#category=kornia.morphology' target='_blank'>Kornia Morphological Operators Tutorial</a></p>"

with gr.Blocks(title=title) as demo:
    gr.Markdown(f"# {title}")
    gr.Markdown(description)
    
    with gr.Row():
        input_image = gr.Image(label="Input Image")
        output_image = gr.Image(label="Output Image")
    
    with gr.Row():
        operator = gr.Dropdown(
            choices=["Dilation", "Erosion", "Open", "Close", "Gradient", "Bottom Hat", "Top Hat"],
            label="Operator"
        )
        kernel = gr.Radio(
            choices=["Ones", "Eye", "Cross"],
            label="Kernel Type"
        )
        kernel_size = gr.Slider(
            minimum=1,
            maximum=7,
            step=2,
            value=3,
            label="Kernel size"
        )
    
    apply_button = gr.Button("Apply Operator")
    apply_button.click(
        fn=morphological_operators,
        inputs=[input_image, operator, kernel, kernel_size],
        outputs=output_image
    )
    
    gr.Examples(
        examples=examples,
        inputs=[input_image, operator, kernel, kernel_size],
        outputs=output_image,
        fn=morphological_operators,
        cache_examples=True
    )
    
    gr.Markdown(article)

if __name__ == "__main__":
    demo.launch()