Skip to content
A creative web application using machine learning model to find shortest path between images
JavaScript CSS Python HTML
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
public
readme
.DS_Store
README.md
features_caltech256.p
graph_caltech256_30knn.p
package-lock.json
recipe.py
requirements.txt
server.js
useownimg.py

README.md

Recipe Recommender

A creative project to suggest a recipe from certain ingredient images. The project is using web scraping, PCA, and kNN techniques and is implemented by Javascript and Python.

User will choose some images of ingredients, the application will give you a cuisine and will show you how to turn the ingredient to the cuisine by image transfer.

Node Server/Client

server.js

Connect the python script and communicate with the client webpage.

public/imageTransfer.js

Receive and send data from/to the server.

public/dropzone.js

The code for Dropzone

Python Script

recipe.py

Get ingredient keywords from server and go online to scrape the cuisine image online.

useownimg.py

Do PCA and kNN in Caltech256 database. Look for the shortest image path between the ingredient image and the cuisine image.

Result

You can’t perform that action at this time.