Extract images from document Leave feedback
On this page
GroupDocs.Parser provides functionality to extract images from various document formats including PDF, Word, Excel, PowerPoint, and more.
- GroupDocs.Parser for Python via .NET installed
- Sample documents containing images
- Write access to save extracted images (optional)
To extract all images from a document:
from groupdocs.parser import Parser
# Create an instance of Parser class
with Parser("./sample.pdf") as parser:
# Extract images
images = parser.get_images()
# Check if image extraction is supported
if images is None:
print("Image extraction isn't supported")
elif not images:
print("No images found")
else:
# Iterate over images
for idx, image in enumerate(images):
# Print image information
print(f"Image {idx + 1}:")
print(f" Page: {image.page.index + 1}")
print(f" Type: {image.file_type}")
print(f" Size: {image.rectangle.size.width}x{image.rectangle.size.height}")
print(f" Position: ({image.rectangle.left}, {image.rectangle.top})")
The following sample file is used in this example: sample.pdf
Image 1:
Page: 1
Type: JPEG Image(.jpeg)
Size: 150x45
Position: (40, 70.99999999593751)
Image 2:
Page: 1
Type: JPEG Image(.jpeg)
Size: 375x234
Position: (110, 383.9999999959375)
[TRUNCATED]
Expected behavior: Returns a collection of PageImageArea objects representing all images found in the document, or None if image extraction is not supported.
To save extracted images to disk:
from groupdocs.parser import Parser
import os
# Create output directory
output_dir = "extracted_images"
os.makedirs(output_dir, exist_ok=True)
# Create an instance of Parser class
with Parser("./sample.docx") as parser:
# Extract images
images = parser.get_images()
if images is None:
print("Image extraction isn't supported")
elif not images:
print("No images found")
else:
# Iterate over images and save them
for idx, image in enumerate(images):
# Get file extension based on image type
extension = image.file_type.extension
# Generate filename
filename = f"image_{idx + 1}{extension}"
filepath = os.path.join(output_dir, filename)
# Save image to file
image.save(filepath)
print(f"Saved: {filepath}")
The following sample file is used in this example: sample.docx
extracted_images/image_1.png (4 KB)
extracted_images/image_2.jpeg (176 KB)
extracted_images/image_3.jpeg (136 KB)
Expected behavior: Saves each extracted image to a separate file with the appropriate file extension (.png, .jpg, .gif, etc.).
To extract images along with detailed metadata:
from groupdocs.parser import Parser
# Create an instance of Parser class
with Parser("./sample.pptx") as parser:
# Check if image extraction is supported
if not parser.features.images:
print("Document doesn't support image extraction")
else:
# Extract images
images = parser.get_images()
if images:
print(f"Found {len(images)} images\n")
for idx, image in enumerate(images):
print(f"Image {idx + 1}:")
print(f" Page: {image.page.index + 1}")
print(f" Format: {image.file_type}")
print(f" Rotation: {image.rotation} degrees")
print(f" Rectangle: {image.rectangle}")
print(f" Width: {image.rectangle.size.width}")
print(f" Height: {image.rectangle.size.height}")
print()
else:
print("No images found")
The following sample file is used in this example: sample.pptx
Found 4 images
Image 1:
Page: 1
Format: Portable Network Graphic(.png)
Rotation: 0 degrees
Rectangle: (30; 20) (150; 45)
Width: 150
Height: 45
[TRUNCATED]
Expected behavior: Displays comprehensive information about each image including position, size, format, and rotation angle.
To work with image data as a stream:
from groupdocs.parser import Parser
from groupdocs.parser.options import ImageOptions, ImageFormat
# Create an instance of Parser class
with Parser("./sample.pdf") as parser:
# Extract images
images = parser.get_images()
if images:
for idx, image in enumerate(images):
# Get image stream (a .NET MemoryStream)
image_stream = image.get_image_stream()
# Read image data: to_array() returns the stream content as bytes
image_data = image_stream.to_array()
print(f"Image {idx + 1}: {len(image_data)} bytes, format: {image.file_type}")
# Optionally convert to PNG
png_options = ImageOptions(ImageFormat.PNG)
png_stream = image.get_image_stream(png_options)
png_data = png_stream.to_array()
print(f" Converted to PNG: {len(png_data)} bytes")
else:
print("No images found")
The following sample file is used in this example: sample.pdf
Image 1: 10365 bytes, format: JPEG Image(.jpeg)
Converted to PNG: 55430 bytes
Image 2: 90027 bytes, format: JPEG Image(.jpeg)
Converted to PNG: 633711 bytes
Image 3: 125180 bytes, format: JPEG Image(.jpeg)
Converted to PNG: 837256 bytes
Image 4: 163435 bytes, format: JPEG Image(.jpeg)
Converted to PNG: 1186275 bytes
Image 5: 94989 bytes, format: JPEG Image(.jpeg)
Converted to PNG: 928726 bytes
Expected behavior: Provides access to raw image data as a stream, with optional format conversion.
To convert images to a specific format during extraction:
from groupdocs.parser import Parser
from groupdocs.parser.options import ImageOptions, ImageFormat
import os
# Create output directory
output_dir = "converted_images"
os.makedirs(output_dir, exist_ok=True)
# Create an instance of Parser class
with Parser("./sample.pdf") as parser:
# Extract images
images = parser.get_images()
if images:
# Create image options for PNG format
png_options = ImageOptions(ImageFormat.PNG)
for idx, image in enumerate(images):
# Save image as PNG (regardless of original format)
filename = f"image_{idx + 1}.png"
filepath = os.path.join(output_dir, filename)
# Save with conversion
image.save(filepath, png_options)
print(f"Saved as PNG: {filepath}")
else:
print("No images found")
The following sample file is used in this example: sample.pdf
converted_images/image_1.png (54 KB)
converted_images/image_2.png (619 KB)
converted_images/image_3.png (818 KB)
converted_images/image_4.png (1158 KB)
converted_images/image_5.png (907 KB)
Expected behavior: All extracted images are converted to PNG format before saving, regardless of their original format.
Extract images from multiple documents:
from groupdocs.parser import Parser
import os
from pathlib import Path
def extract_images_from_directory(input_dir, output_dir):
"""
Extract images from all documents in a directory.
"""
os.makedirs(output_dir, exist_ok=True)
# Supported document extensions
extensions = ['.pdf', '.docx', '.doc', '.xlsx', '.pptx', '.ppt']
for file_path in sorted(Path(input_dir).rglob('*')):
if file_path.suffix.lower() in extensions:
print(f"\nProcessing: {file_path.name}")
try:
with Parser(str(file_path)) as parser:
images = parser.get_images()
if images is None:
print(f" Image extraction not supported")
continue
# Create subdirectory for this document
doc_output_dir = os.path.join(output_dir, file_path.stem)
os.makedirs(doc_output_dir, exist_ok=True)
# Save images
image_count = 0
for idx, image in enumerate(images):
filename = f"image_{idx + 1}{image.file_type.extension}"
filepath = os.path.join(doc_output_dir, filename)
image.save(filepath)
image_count += 1
print(f" Extracted {image_count} images")
except Exception as e:
print(f" Error: {e}")
# Usage
extract_images_from_directory("documents", "extracted_images")
The following sample files are used in this example: annual-report.pdf, product-catalog.pptx, service-agreement.docx
extracted_images/annual-report/image_1.jpeg (10 KB)
extracted_images/annual-report/image_2.jpeg (88 KB)
extracted_images/annual-report/image_3.jpeg (122 KB)
extracted_images/annual-report/image_4.jpeg (160 KB)
extracted_images/annual-report/image_5.jpeg (93 KB)
extracted_images/product-catalog/image_1.png (4 KB)
extracted_images/product-catalog/image_2.jpeg (136 KB)
extracted_images/product-catalog/image_3.jpeg (75 KB)
extracted_images/product-catalog/image_4.jpeg (105 KB)
extracted_images/service-agreement/image_1.png (4 KB)
[TRUNCATED] (12 files total)
- The
get_images()method returnsNoneif image extraction is not supported for the document format - Always check
parser.features.imagesbefore attempting to extract images - Supported output formats: BMP, GIF, JPEG, PNG, WebP
- Images are extracted with their original quality and resolution
- The
rotationproperty indicates the image rotation angle (0, 90, 180, or 270 degrees) - Use
get_image_stream()for memory-efficient processing of large images
Was this page helpful?
Any additional feedback you'd like to share with us?
Please tell us how we can improve this page.
Thank you for your feedback!
We value your opinion. Your feedback will help us improve our documentation.
