Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/trykatchup/poke-pi-dex
Our deep learning for computer vision related project for nostalgic poke weebs (Sistemi digitali, Unibo).
https://github.com/trykatchup/poke-pi-dex
artificial-intelligence cnn cnn-keras computer-vision deep-learning image-classification image-processing machine-learning numpy opencv poke-pi-dex pokedex pokemon pygame raspberry-pi-4 raspberrypi tensorflow tkinter
Last synced: 2 months ago
JSON representation
Our deep learning for computer vision related project for nostalgic poke weebs (Sistemi digitali, Unibo).
- Host: GitHub
- URL: https://github.com/trykatchup/poke-pi-dex
- Owner: TryKatChup
- License: gpl-3.0
- Created: 2021-07-29T14:23:53.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2024-02-11T12:02:57.000Z (11 months ago)
- Last Synced: 2024-10-13T05:23:07.403Z (2 months ago)
- Topics: artificial-intelligence, cnn, cnn-keras, computer-vision, deep-learning, image-classification, image-processing, machine-learning, numpy, opencv, poke-pi-dex, pokedex, pokemon, pygame, raspberry-pi-4, raspberrypi, tensorflow, tkinter
- Language: Jupyter Notebook
- Homepage:
- Size: 102 MB
- Stars: 22
- Watchers: 2
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.it.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Roadmap: docs/roadmap.md
Awesome Lists containing this project
README
[![Contributors][contributors-shield]][contributors-url]
[![Forks][forks-shield]][forks-url]
[![License][license-shield]][license-url]
[![Stargazers][stars-shield]][stars-url]
[![Downloads][downloads-shield]][downloads-url]
[![Issues][issues-shield]][issues-url]
[![Poke-Pi-Dex][poke-pi-dex-shield]][poke-pi-dex-url]
[![Raspberry Pi][raspberry-shield]][raspberry-url]
[![Python][python-shield]][python-url]
[![nVIDIA][nvidia-shield]][nvidia-url]
[![Keras][keras-shield]][keras-url]
[![TensorFlow][tensorflow-shield]][tensorflow-url]
[![Open CV][opencv-shield]][opencv-url]
Poké-Pi-Dex
**Poké-Pi-Dex** è il nostro progetto per poké-weeb nostalgici, basato su deep learning / computer vision. Realizzato da [Karina Chichifoi](https://github.com/TryKatChup) e [Michele Righi](https://github.com/mikyll).
Abbiamo ricreato il clone di un **Pokédex** che riconosce immagini di Pokémon della prima generazione, sfruttando una Rete Neurale Convoluzionale. È stato sviluppato per eseguire su un **Raspberry Pi4** con display LCD, PiCamera ed altri componenti collegati.
Il case è fatto di cartoncino riciclato. 🌱
Relazione
·
Presentazione
·
Att. Progettuale
·
Richiedi una Feature|Segnala un Bug
·
English 
Indice
## Demo
Guarda la [demo](https://youtu.be/IkbLYq1PmRs) su YouTube!
Demo video
## Funzionalità
TODO
## Utilizzo
To use the application follow these steps:
### Prerequisiti
- conda
TO-DO
### Installazione
#### Windows
1. Scarica l'[ultima versione](https://github.com/TryKatChup/Poke-Pi-Dex/releases/latest)
2. Estrai lo zip
3. Crea il virtual environment:
```bash
conda env create -f environment.yml
```
4. Esegui l'applicazione:
```bash
python poke-pi-dex.py
```
## Roadmap
- [x] Dataset
- [x] trovare un dataset adatto per la rete neurale
- [x] sistemarlo (ritagliare le immagini) ed estenderlo
- [x] Classificatore
- [x] CNN con 3 layer convoluzionali e 2 layer FC
- [x] data augmentation (specchiamento, rotazione, contrasto e ~~luminosità~~ randomici)
- [x] provare dropout
- [x] provare batch norm
- [x] grafici loss e accuracy
- [x] test con immagini reali
- [x] miglirare la vecchia CNN
- [x] Applicazione
- [x] Repository Pokémon
- [x] trovare un file .json e caricarlo in un dizionario
- [x] controllarlo e sistemarlo
- [x] creare una classe Pokémon
- [x] input video
- [x] creare una classe separata
- [x] creare una funzione che permette di ottenere un frame dalla PiCamera (e fare i test)
- [x] visualizzare l'immagine all'interno di un canvas
- [x] struttura GUI
- [x] creare un menu principale
- [x] creare un pannello delle informazioni sull'app
- [x] creare una schermata per il Pokédex, divisa in 2 parti (sinistra per il video input, destra per i dettagli del Pokémon)
- [x] creare una vista per le impostazioni
- [x] bottone per ottenere il frame corrente
- [x] etichette ed entry per i dettagli del Pokémon (statistiche con barre dinamiche e di colori differenti)
- [x] aggiungere bottoni per scorrere fra le evoluzioni successive (ad esempio: Eevee ha diverse evoluzioni possibili)
- [x] cambiare la entry del "tipo/i" (da testo a immagine)
- [x] aggiungere bottone per riprodurre il verso
- [x] raccogliere i file audio dei versi
- [x] aggiungere la lettura della descrizione
- [x] ottenere i file audio delle descrizioni utilizzando un bot di lettura
- [x] realizzare l'aggiornamento per lingue differenti
- [x] rendere le impostazioni persistenti
[downloads-shield]: https://img.shields.io/github/downloads/TryKatChup/Poke-Pi-Dex/total
[downloads-url]: https://github.com/TryKatChup/Poke-Pi-Dex/releases/latest
[contributors-shield]: https://img.shields.io/github/contributors/TryKatChup/Poke-Pi-Dex
[contributors-url]: https://github.com/TryKatChup/Poke-Pi-Dex/graphs/contributors
[forks-shield]: https://img.shields.io/github/forks/TryKatChup/Poke-Pi-Dex?style=flat
[forks-url]: https://github.com/TryKatChup/Poke-Pi-Dex/network/members
[stars-shield]: https://img.shields.io/github/stars/TryKatChup/Poke-Pi-Dex?style=flat
[stars-url]: https://github.com/TryKatChup/Poke-Pi-Dex/stargazers
[issues-shield]: https://img.shields.io/github/issues/TryKatChup/Poke-Pi-Dex
[issues-url]: https://github.com/mikyll/TryKatChup/Poke-Pi-Dex/issues
[license-shield]: https://img.shields.io/badge/License-GPLv3-blue.svg
[license-url]: http://perso.crans.org/besson/LICENSE.html
[ask-me-anything-shield]: https://img.shields.io/badge/Ask%20me-anything-1abc9c.svg
[ask-me-anything-url]: https://github.com/TryKatChup/Poke-Pi-Dex/issues
[open-collab-shield]: https://colab.research.google.com/assets/colab-badge.svg
[open-collab-url]: https://github.com/TryKatChup/Poke-Pi-Dex/issues
[made-with-phyton-shield]: https://img.shields.io/badge/Made%20with-Python-14354C.svg
[made-with-phyton-url]: https://www.python.org/
[made-with-markdown-shield]: https://img.shields.io/badge/Made%20with-Markdown-1f425f.svg
[made-with-markdown-url]: http://commonmark.org
[open-source-shield]: https://badges.frapsoft.com/os/v1/open-source.png?v=103
[open-source-url]: https://github.com/ellerbrock/open-source-badges/
[poke-pi-dex-shield]: https://custom-icon-badges.herokuapp.com/badge/pok%C3%A9pidex-wow-orangered?logo=poke-pi-dex
[poke-pi-dex-url]: https://github.com/TryKatChup/Poke-Pi-Dex
[raspberry-shield]: https://img.shields.io/badge/-RaspberryPi-C51A4A?&logo=Raspberry-Pi
[raspberry-url]: https://www.raspberrypi.org/
[keras-shield]: https://img.shields.io/badge/Keras-%23D00000.svg?logo=Keras&logoColor=white
[keras-url]: https://keras.io/
[tensorflow-shield]: https://img.shields.io/badge/TensorFlow-%23FF6F00.svg?logo=TensorFlow&logoColor=white
[tensorflow-url]: https://www.tensorflow.org/
[opencv-shield]: https://img.shields.io/badge/opencv-%23white.svg?logo=opencv&logoColor=white
[opencv-url]: https://opencv.org/
[nvidia-shield]: https://img.shields.io/badge/nVIDIA-%2376B900.svg?logo=nVIDIA&logoColor=white
[nvidia-url]: https://www.nvidia.com/
[python-shield]: https://img.shields.io/badge/python-3670A0?logo=python&logoColor=ffffff
[python-url]: https://www.python.org/