https://github.com/thecoderpinar/-space-image-classificatios
Classification of Space Images: Space Science Analysis with Artificial Intelligence
https://github.com/thecoderpinar/-space-image-classificatios
Last synced: 10 months ago
JSON representation
Classification of Space Images: Space Science Analysis with Artificial Intelligence
- Host: GitHub
- URL: https://github.com/thecoderpinar/-space-image-classificatios
- Owner: ThecoderPinar
- License: mit
- Created: 2023-06-14T20:49:14.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2023-06-14T20:57:10.000Z (almost 3 years ago)
- Last Synced: 2025-04-03T15:28:27.723Z (about 1 year ago)
- Language: Jupyter Notebook
- Size: 1.96 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Space Image Classificatios 🚀
Classification of Space Images: Space Science Analysis with Artificial Intelligence
# Project Title 🔥
This project aims to create a model using Python and artificial intelligence techniques to classify space images. The project utilizes space images obtained from NASA's public data archive or other sources.
In the first step, a suitable dataset is selected or created. For instance, a dataset consisting of images captured by the Hubble Space Telescope can be used. The dataset should include different space objects (galaxies, star clusters, planets, etc.) and diverse space scenes.
Next, the images are opened and visualized using Python and relevant libraries. The images are resized to appropriate dimensions, color channels are normalized, and preprocessing steps like noise reduction or normalization are performed.
The ResNet50 model, a popular deep learning model, is employed as the artificial intelligence model. The pretrained ResNet50 model is used for image classification. The images are reshaped to match the input dimensions of the model and converted into a format understandable by the model. Subsequently, the model classifies the images and obtains the highest prediction.
The results are visualized and displayed on the screen. The predicted class name and confidence value are presented along with the image. This allows observation of the predicted object or scene for the space images and the level of confidence in the model's predictions.
This project helps improve your skills in Python programming, image processing, and deep learning techniques. It also serves as an interesting example for individuals interested in space science or scientific research.
# Before Image
# After Image
