Is it possible to automate the input of a string of data from a physical object into a file?

E

Estupido2020-05-02 00:05:15

Image processing

Estupido, 2020-05-02 00:05:15

Please help me decide how to automate the task.
There are 6k pieces of packages that have stickers with a code on them. View code A12345678B.
These codes must be included in the file.
It is proposed to manually enter codes by a person from photographs.
What options can be considered for saving the string with the code to a file at the time of photographing the stickers from the packages?
What are the options for recognizing 6k+ approximately uniform images on a PC?
Is something like a CCP scanner able to read such a line?

This is a colorful package of a drug, corresponding to the name of the drug in the picture, and foil stamping, and of the sticker itself. The picture is simple, but colorful for ocr.

Reply

Answer the question

In order to leave comments, you need to log in

2 answer(s)

D

Dimonchik, 2020-05-02
@dimonchik2013

https://ru.wikipedia.org/wiki/Tesseract
and specifically for coding, for example, pytesseract - there are wrappers for any popular language

R

Roman Mirilaczvili, 2020-05-02
@2ord

This is a colorful package of a drug, corresponding to the name of the drug in the picture, and the foil stamping, and of the sticker itself. The picture is simple, but colorful for ocr.

It is possible to automate recognition, in theory. The result is highly dependent on the quality of the image. Depending on this, the percentage of errors in recognition can be both small and large.
Technically, there are two options:

industrial scanner capable of character recognition. Search the Internet: pen scanner ocr . Such options, in theory, are capable of transmitting a code to a computer via Bluetooth, USB. Perhaps one by one into an open Excel spreadsheet or directly into a file with a list. This is the shorter and preferred route for business.
if you want to do it yourself “on your knee”, then shoot mobile phones at the right angle so that there are as few glare as possible, then transfer them to a PC / laptop, where, after pre-processing the images, perform code recognition with specialized software for OCR or programmatically (Tesseract and others ) and save them to a file or somewhere else.