Skip to content

Text detector

The text is extracted from the images using google-cloud-vision. For this, you need an API key. Set up your google account following the instructions on the google Vision AI website or as described here. You then need to export the location of the API key as an environment variable: 

export GOOGLE_APPLICATION_CREDENTIALS="location of your .json"
The extracted text is then stored under the text key (column when exporting a csv).

Googletrans is used to recognize the language automatically and translate into English. The text language and translated text is then stored under the text_language and text_english key (column when exporting a csv).

The English text is cleaned from numbers and unrecognized words (text_clean).

Key Features

  • Text Extraction: Detects and extracts text from images using Google Cloud Vision API
  • Language Detection: Automatically detects the language of extracted text
  • Translation: Translates text to English using Google Translate
  • Text Preprocessing:
  • Handles formatting issues (adds spaces after periods for translation compatibility)
  • Truncates very long texts (default: 5000 characters)
  • Removes line breaks
  • NLP Analysis: Integrates with spaCy for advanced text analysis
  • CSV Support: TextAnalyzer class for reading text from CSV files
  • Privacy Controls: Requires explicit privacy disclosure acceptance before processing

Usage

detector = ammico.TextDetector(
    subdict={"filename": "image.jpg"},
    skip_extraction=False,  # Set True if text already provided
    accept_privacy="PRIVACY_AMMICO"
)
results = detector.analyse_image()
It is also possible to read in a csv file for translation and processing: 
detector = ammico.TextDetector(
    csv_path="./text_input.csv",
    column_key="Text column",
    csv_encoding="utf-8",
    skip_extraction=True,  
    accept_privacy="PRIVACY_AMMICO"
)
results = detector.analyse_image()
In this case, the extraction of the text from the image is skipped and the text is translated directly (no Google Cloud Vision API key required).

Output

Returns dictionary with:

  • text: Original extracted text
  • text_language: Detected source language code
  • text_english: Translated English text
  • text_truncated: Truncated version if original exceeds length limit

Requirements

  • Google Cloud Vision API credentials
  • Privacy disclosure acceptance (via environment variable or interactive prompt)
  • spaCy English model (en_core_web_md)

Workflow

flowchart TD
    Start([Start]) --> Init[Initialize TextDetector]
    Init --> CheckPrivacy{Privacy Accepted?}

    CheckPrivacy -- No --> Error[Raise ValueError]
    CheckPrivacy -- Yes --> CheckSkip{Skip Extraction?}

    CheckSkip -- No --> GetText[get_text_from_image]
    GetText --> CloudVision[Call Google Cloud Vision API]
    CloudVision --> CheckResp{Response?}
    CheckResp -- Error --> LogError[Log Error]
    CheckResp -- Success --> Extract[Extract Text]

    CheckSkip -- Yes --> ReadDict[Read text from subdict]

    Extract --> HasText{Text Found?}
    ReadDict --> HasText

    HasText -- No --> End([End])
    HasText -- Yes --> Preprocess[Preprocess Text]

    Preprocess --> AddSpc[_check_add_space_after_full_stop]
    AddSpc --> Trunc[_truncate_text]
    Trunc --> Trans[translate_text]
    Trans --> GTrans[Call Google Translate]
    GTrans --> Clean[remove_linebreaks]

    Clean --> RunSpacy[_run_spacy]
    RunSpacy --> RetDict[Return Updated Dictionary]
    RetDict --> End