RedCarpet - Backend

This github repo contains the files needed to deploy the Digital Ocean Back-End entity that will handle requests from the RedCarpet Mobile Application.

Requirements

In order to deploy the backend implementation, follow these steps:

Instantiate a cloud server of your choice that is running on the Ubuntu Operating System.

Install Python3 and Python pip using:

    sudo apt-get update
    sudo apt-get install python3
    sudo apt-get install python3-pip

Install the necessary Python Modules using pip. The modules are listed in the file "pip_modules.txt"

Currently, the implemented files are as follows:

1. palette_server.py

Palette Server

This file contains the necessary request handlers to perform the reverse image searching. Users will upload a picture through the post method under the path "/api/classify/". The corresponding picture will be piped into through the function "goog_cloud_vision", which returns the top ten labels and top ten logos associated with the contents of the picture in a dictionary of strings. We are using the Google Cloud Vision API with our special API key to make this possible. This returned dictionary will then be passed in as an argument in to the function "searchParses". This function takes the dictionary, and parses the keywords and returns the top five search results from Microsoft's Bing Search Engine via HTTPS request. The server will return the URL's in a json format to the frontend, where the application will display the relevant results to the user.