https://github.com/sahil210695/convolution-evolution

This is to see how a kernel will convolve over an image and what will be its output after convolution
https://github.com/sahil210695/convolution-evolution

cnn convolutional-neural-networks data-science image-kernel image-processing numpy numpy-neural-network python streamlit

Last synced: 3 months ago
JSON representation

This is to see how a kernel will convolve over an image and what will be its output after convolution

• Host: GitHub
• URL: https://github.com/sahil210695/convolution-evolution
• Owner: sahil210695
• Created: 2020-07-27T15:54:55.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2020-07-27T16:08:17.000Z (almost 5 years ago)
• Last Synced: 2025-01-21T11:45:37.335Z (5 months ago)
• Topics: cnn, convolutional-neural-networks, data-science, image-kernel, image-processing, numpy, numpy-neural-network, python, streamlit
• Language: Python
• Homepage:
• Size: 15.6 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Convolution is Evolution 😎

### This is to see how a kernel will convolve over an image and what will be its output after convolution

#### Kernels that we are using:

- **Horizontal** kernel will extract all *horizontal* edges

- **Vertical** kernel will extract all *vertical* edges

- **Edge** kernel will extract all *edges*

#### Demo Image

![Demo Image](assets/demo.png)

#### Kernels with their output:

1. Horizontal Kernel

```python

np.array([

    [-1, -1, -1],

    [ 0,  0,  0],

    [ 1,  1,  1],

])

```

![Horizontal Kernel](assets/h_krnl_output.jpg)

2. Vertical Kernel

```python

np.array([

    [-1,  0,  1],

    [-1,  0,  1],

    [-1,  0,  1],

])

```

![Vertical Kernel](assets/v_krnl_output.jpg)

3. Edge Kernel

```python

np.array([

    [-1, -1, -1],

    [-1,  8, -1],

    [-1, -1, -1],

])

```

![Edge Kernel](assets/e_krnl_output.jpg)

#### UI

I have created UI for this demo using **[Streamlit](https://www.streamlit.io/)**

- `pip install streamlit`

- `streamlit run app.py`