varun-singh14/image-to-text

Image to Text Conversion

Image Processing Project Report

Introduction

This project aims to separate text and visual elements from an image using a Vision API and basic image segmentation techniques.

Technologies Used

• Google Cloud Vision API: For text extraction (OCR).
• OpenCV: For visual element segmentation.
• Python: For scripting and automation.

Implementation Details

Text Extraction

The text_extraction.py script uses Google Cloud Vision API to extract text from an image. The text is retrieved along with its coordinates.

Visual Element Segmentation

The visual_segmentation.py script uses OpenCV to detect and isolate visual elements. Techniques like edge detection and contour detection are employed.

HTML Generation

The html_generator.py script converts the extracted text and visual elements into a basic HTML structure, embedding text within <p> tags and visual elements within <img> tags.

Challenges Encountered

• OCR Accuracy: Handling OCR inaccuracies and noise in the text extraction process.
• Visual Element Detection: Isolating meaningful visual elements while filtering out noise and irrelevant parts.

Conclusion

This project successfully separates text and visual elements from images and generates an HTML representation. Future improvements could include advanced layout handling and more robust visual element detection.