High-Level Project Summary
NASA has a large database of photos, sometimes it is difficult to find that one specific photo.We developed mobile application which allows to enter your text and on submit returns most relevant image to these words.
Link to Final Project
Link to Project "Demo"
Detailed Project Description
How is work:
- The user enters the text (it can be several sentences) according to which he wants to find a suitable photo.
- The entered text is processed with the help of NLP techniques and keywords are selected
- A new url is generated with the received keyword that returned specific photos
Benefits:
- It helps to find a photo more easily, not based on a few words, but a short text of a few sentences
- Easier images categorization/classification and add metadata
Technologies:
Frontend:
- Flutter
- Dart
Backend
- Python
- Django
Search Engine:
- NLTK
- Google Colab
Space Agency Data
Space agency NASA collects Nasa image of day (https://www.nasa.gov/multimedia/imagegallery/iotd.html),we using this NASA api https://images-api.nasa.gov/
Hackathon Journey
We found out that NASA really has large amounts of photo data and that they are not used. We see in what format they are stored and how they are described. We see the possibilities of using photos to create added value.
References
Data: https://www.nasa.gov/multimedia/imagegallery/iotd.html, API https://images-api.nasa.gov/
Code: Flutter, Django, colab
NLP: punctuation,word tokeniztion,POS tag, Name Entity Recognition
Tags
nasa search engine ai nlp