Python Script to process message content (with image adjustment and OCR)

Completado Publicado hace 3 años Pagado a la entrega
Completado

This project is to create a python script that will (1) download messages from a Discord feed and any full-sized images (in links), image adjust those images, use an OCR step and output the result into a CSV file.

This project outline:

1) Have a simple program settings file which holds a token ID, feed ID and date/time of the last read message.

2) Using [login to view URL] download a discord feed (one message at a time).

3) For any pictures in the current message (usually a URL to a PNG file) then download them at large (original) size to an 'images' sub-folder.

4) Using Python Pillow / PIL library (or similar) to filter the image to retain only pixels with 0a5394 (hex) RGB value. Everything else is deleted (left as white). One way to do this (if the library allows) is to simply select everything by that colour and paste the result to a new image file.

5) Using Python Tesseract library (or similar) Ref: [login to view URL] convert the image to text.

6) Repeat 4 and 5 but this time retain only pixels with 0a000000 (hex) RGB value (black). Fill the rest of the image with black. Invert the image colour (black becomes white).

7) CSV output file is in this format. Date/Time, Message ID (if available), Post text, URL link to image, filename of locally saved original image (exclude path), OCR values in step 5 (separated by | (vertical line symbol) for each OCR line), OCR values in step 6 (seperated by | for each line).

8) Update the setting file to show the current post that has been processed (so next time the script is run it will only download new messages after this message ID or time/date)

The final script will be executed every few minutes (for about 3 hours per day) to check on the Discord feed and retrieve new messages.

For development you will need to create a discord account for testing. I cannot unfortunately provide access to the actual feed due to it's a private channel but you can use any other feed to test the program logic of retrieving messages and updating the setting.

For steps 3-7 I will be providing 4-5 example images to work with so can test and make sure the steps are working okay.

Please PM me if you'd like to see the example images first and some other detail.

In you offer either make it a fixed price of give me a real indication for the number of hours for this work so I have some idea of budget.

Code should be clean and tidy and easy to follow. Please use version numbering for each version being sent.

Python

Nº del proyecto: #29096117

Sobre el proyecto

11 propuestas Proyecto remoto Activo hace 3 años

Adjudicado a:

kevinlee1238

Hello, sir I am a professional OCR developer. I developed several products for image processing [login to view URL] [login to view URL] We can discuss the project o Más

$12 USD / hora
(10 comentarios)
4.7

11 freelancers están ofertando un promedio de $15 / hora por este trabajo

talesmoraes2001

Hello. Thanks for your job posting. I read your description carefully and understand sufficiently. I have 6 years experience for computer vision and Machine Learning. I have a lot of projects such as face detection and Más

$10 USD / hora
(8 comentarios)
3.6
valerykrapiv

Hello Dear, Keep safe and protect yourself from virus I read with interest your proposal entitled. I am a python OCR developer having experience of 7 years. If you are interested in my quote, Please contact me. Look Más

$15 USD / hora
(1 comentario)
3.1
ginushaji3

Hi, I have read the job description. I have strong hands on in Python with more than 6 years of experience. Please contact me for further proceedings. Warm regards, Ginu

$13 USD / hora
(2 comentarios)
2.6
arbitratorva

@@@@ML&DL&Computer Vision Expert@@@@ Thanks for your posting! I am a computer vision and machine learning expert with full experiences in tensorflow, darknet, keras, pytorch, opencv and open vino, android, iOS, C#, C++ Más

$15 USD / hora
(2 comentarios)
2.0
ArastuSystems1

Hey there, Acc. to description, you want us to download a discord feed(one message at a time), for each message , we need to download the picture at their original size and store them in " images" sub folder . We nee Más

$13 USD / hora
(1 comentario)
1.4
Niosmart

Hi we are python expert We can do your work as per your requirement message us to discuss in chat Thanks Somya

$30 USD / hora
(0 comentarios)
0.0