How to extract the data from a electrical circuit diagram using Microsoft Document intelligence

Hello @Ajeet Ojha

Thank you for your response.

Solution

To extract paragraphs from the JSON output provided by Azure AI Document Intelligence, you'll need to parse the JSON and reconstruct the paragraphs from the individual words. I put together a step-by-step guide and corresponding Python code to help you achieve this.

1. Set Up Your Environment:

• Ensure you have Python installed on your laptop.
• Install necessary libraries using pip: pip install requests

2. Get JSON Output from Azure Document Intelligence: This example assumes you already have the JSON output saved to a file called output.json.
3. Parse the JSON and Reconstruct Paragraphs:

• Load the JSON file.
• Iterate through the JSON structure to extract words and their positions.
• Reconstruct paragraphs based on word positions and sequences.

The example of python code that can perform the number 3 above:

import json
def extract_paragraphs(json_data):
    # This function extracts paragraphs from the Azure Document Intelligence JSON output
    paragraphs = []
    for page in json_data['analyzeResult']['readResults']:
        for line in page['lines']:
            paragraph = " ".join([word['text'] for word in line['words']])
            paragraphs.append(paragraph)
    return paragraphs
def main():
    # Load JSON data from a file
    with open('output.json', 'r') as file:
        json_data = json.load(file)
    
    # Extract paragraphs
    paragraphs = extract_paragraphs(json_data)
    
    # Print extracted paragraphs
    for i, paragraph in enumerate(paragraphs):
        print(f"Paragraph {i+1}: {paragraph}\n")
if __name__ == "__main__":
    main()

This is an explanation and things you will need to do:

The main function opens and reads the JSON data from output.json. Also, the extract_paragraphs function processes the JSON data. It assumes that each page contains lines, and each line contains words. It concatenates words to form lines (which are treated as paragraphs here). Then, the paragraphs are printed to the console.

What to do to make the script work:

1. Save the JSON output from Azure Document Intelligence to a file named output.json.
2. Save the Python script above to a file, e.g., extract_paragraphs.py.
3. Run the script from the command line: python extract_paragraphs.py

Please kindly take notes:

• Adjust the extract_paragraphs function if the JSON structure differs.
• You might need additional parsing logic if your JSON has a more complex structure or if you need to differentiate between different types of text elements (e.g., paragraphs, tables, etc.) or generate output to another file format.

References

From the additional resources by the right side of this page you learn or read more on the above solution using the links below:

Extract data from forms with Azure Document intelligence - Training

Accept Answer

I hope this is helpful! Do not hesitate to let me know if you have any other questions.

** Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful ** so that others in the community facing similar issues can easily find the solution.

Best Regards,

Sina Salam