二手鞋資訊轉換平台 ( The Information Transformation Platform of Second-Hand Shoes )

Since second-hand trading becomes a current trend, there are more and more people would like to sell their own object as a second-hand good. Taking the existing shopping platforms as the examples, the sellers have to fill in all the informations about their goods by themself, which is inconvenient sometimes. Our goal is to simplify the listing process of selling second-hand shoes, and to promote the willing of putting goods on the shelf. In this project, the Info Transformation Platform can return the depreciation rates, brands, styles, and web searching results of the shoes by taking or uploading photos from users.

Build with Windows OS and Azure.

📶 Introduction by Powerpoint

🎦 DEMO video

Files

DIR //computer_vision//

/evaluate/

model training programs

/model/

trained-model

/img_web_crawler.py

for web crawling

/main.py

main program of the computer vision: to detect the brands of the input img

/size.py

take a picture by webcam and analysize the output image

Libraries

Usage (only for Computer Vision)

1. clone the repo

git clone https://github.com/VivianChan1998/Old-shoes.git

2. install dev dependencies

install and activate virtual environment (https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/)

install pip

py -m pip --version

install virtual environment

py -m pip install --user virtualenv

create virtual environment: in the folder of this project

py -m venv env

activate the virtual environment that have been built

.\env\Scripts\activate

install python packages

pip install requests matplotlib pillow opencv-python

3. register the Azure portal and set up your own computer vision resource(https://portal.azure.com/)

4. Create two directories: "JSON_file" & "img" in the proper site of your computer

5. set up the "config_own.py" file

# First create the "config_own.py" file in the "computer_vision" directory
# and then set up the parameters: JSON_DIR, SHOT_CV2_DIR, IMG_PATH, IMG_DIR

JSON_DIR = "your-own-path" # which stands for the path of the JSON directory
SHOT_CV2_DIR = "your-own-path" # which stands for the path of the images that cv2 camera shoot
IMG_PATH = "your-own-path" # which stands for the path of the to-be-analysized-image
IMG_DIR = "your-own-path" # which stands for the path of the directory where the to-be-analysized-images store

6. in the project directory, type the following command and execute

python3 main.py <your-cv-sunscription-key> <your-cv-endpoint>