Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/sayannath/pothole-detection
Potholes detection model with CNN.
https://github.com/sayannath/pothole-detection
cnn colab deep-learning h5 image-classification keras neural-network tensorflow2 tflite
Last synced: 14 days ago
JSON representation
Potholes detection model with CNN.
- Host: GitHub
- URL: https://github.com/sayannath/pothole-detection
- Owner: sayannath
- License: mit
- Created: 2020-07-11T17:39:31.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2021-04-01T23:22:27.000Z (over 3 years ago)
- Last Synced: 2023-03-04T05:15:00.798Z (over 1 year ago)
- Topics: cnn, colab, deep-learning, h5, image-classification, keras, neural-network, tensorflow2, tflite
- Language: Jupyter Notebook
- Homepage:
- Size: 11.6 MB
- Stars: 5
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Pothole-Detection
## Short Description
A deep learning model to detect potholes.
# Dataset:
Dataset Link:- "https://www.kaggle.com/atulyakumar98/pothole-detection-dataset"
### 👉 If you like this repo then please give it a ⭐️
## NOTE:
This project is subject to change by the developer and is still in development
## To test this out
Testing this is as simple as it gets. Follow this steps:
1. Clone this project.
2. Load the .h5 file in Google Colab
3. Run the following code
```
import numpy as np
from google.colab import files
from keras.preprocessing import image
uploaded=files.upload()
for fn in uploaded.keys():
# predicting images
path='/content/' + fn
img=image.load_img(path, target_size=(64, 64))
x=image.img_to_array(img)
x=np.expand_dims(x, axis=0)
images = np.vstack([x])
classes = model.predict(images, batch_size=32)
print(classes)
if classes[0]>0:
print(fn + " is a pathole")
else:
print(fn + " is a normal")
```
4. When the upload is prompt, choose the image from test_image directory or you can have your own image.
## Pull Request
Pull Requests are welcome. Please follow these rules for the ease of understanding:
* Make sure to check for available issues before raising one
* Give me a maximum of 24-48 hours to respond
* Have proper documentation on the parts you are changing/adding
## Developed & Maintained by
[👨 Sayan Nath](https://sayannath.biz)
[📷 Insta](https://www.instagram.com/sayannath235)
[🐤 Twitter](https://twitter.com/SayanNa20204009)
[🧳 LinkedIn](https://www.linkedin.com/in/sayan-nath-15a989182)
[](https://github.com/sayannath)