rajsanjib/image-processing

Categorize and Extract data from various image documents

Text Detection on Image Using OpenCV and pytesseract

1. Clone the repo

git clone [email protected]:covideoin/image-processing.git or git clone https://github.com/covideoin/image-processing

2. Inside the image-processing directory

Create a virutalenv

virtualenv env -p python3

Activate virutalenv

source env/bin/activate

Install Dependencies

pip install -r requirements.txt

**Might need to install following dependency for tensorflow

sudo apt-get install libcudart10.1

Setup Flask application with gunicorn and nginx

https://www.digitalocean.com/community/tutorials/how-to-serve-flask-applications-with-gunicorn-and-nginx-on-ubuntu-18-04

Install dlib

Try installing python module of dlib pip install dlib

wget http://dlib.net/files/dlib-19.6.tar.bz2
tar xvf dlib-19.6.tar.bz2
cd dlib-19.6/
mkdir build
cd build
cmake ..
cmake --build . --config Release
sudo make install
sudo ldconfig

apis

api is served on port 5000

• /extract_pan_card

  • Request Method [POST]
  • Request

  {"image": "base64 encoded image"}

  • Response

{
  data: {
    "DOB": "",
    "Fathers Name": "",
    "Name": "",
    "pan-id": ""
  },
  signature: "base64 data"
}

• /extract_adhar_card

  • Method [POST]
  • Request

  {"image": "base64 encoded image"}

  • Response

  data: {
    "DOB": "",
    "Name": "",
    "Gender": "",
    "adhar-id": ""
  },
  

  or

  data: {
      "address": address
  }
  

  if backside is given

• /compare_faces

  • Method [POST]
  • Request

      {
       'image1': 'base64 encoded image',
       'image2': 'base64 encoded image'
      }
      ```
  
  - Response
    ```
      {
        score: match_value[0-1]
      }
    ```
  
  

• /convert_pdf

  • Method [POST]
  • Request

      {
       'file': 'base64 encoded pdf file',
      }
      ```
  
  - Response
    ```
      {
        "image_filename": "filename.jpg"
      }
    ```
  

converted image will be stored as temp/pdf-to-image/filename.jpg

• /classify

  Returns predicted label of given document ==> pan-card | aadhar-card | driving-license
  • Method [POST]
  • Request

      {
       'image': 'base64 encoded image file',
      }
      ```
  
  - Response
    ```
      {
        "label": "pan-card/aadhar-card/driving-license"
      }
    ```