Extract Photo & Signature from Scanned PDF Files (Python / Image Processing)

@JaibhanSinghGaur

Hi, I’ve carefully reviewed your requirements and understand this is image-based PDF processing, not OCR-only. The goal is to reliably extract photo and signature images from scanned PDFs, even when layouts shift slightly. I can deliver a Python-based solution using OpenCV + PyMuPDF/pdf2image + NumPy with: Proper DPI handling (300–400) and deskew/pre-processing ROI-based detection with configurable JSON settings Fallback face detection if photo box detection fails Signature extraction using contour/ink-based image processing Batch folder processing with consistent filename outputs Clean, reusable, well-structured code with logging Debug images for failed cases and multi-page fallback logic I have hands-on experience with scanned document processing and can deliver working, production-ready logic, not a demo script. I’m happy to iterate using your sample PDFs and feedback. Let’s discuss timelines and expected accuracy, and I can start immediately. Best regards, Jaibhan

@larsenm2020

With a passion for learning and a proven track record in Python and JavaScript, I am ideally placed to tackle the challenges your projects presents. Even better, I have extensive experience with image processing and using libraries such as OpenCV, PyMuPDF, and NumPy, all of which are vital components of developing the reliable script/system you need. And although my skills broaden beyond image processing, that precise background is exactly why you should choose me. Understanding the unique nature of scanned documents instead of relying only on OCR is key to overcoming the project's hurdles. More than simply delivering "demo scripts," I will provide a clean, reusable, well-structured code that will meet each of your requirements, including those additional "Nice to Have (Bonus)" ones. But don't just take my word for it; I'll keep you involved throughout the process, implementing your feedback as we go. Post completion I can even provide explanatory documentation so future users can maintain and build upon it themselves.

@amitt800

I can create a python script that will extract photo and signature from those pdf files. I would like to see the few those PDF file, so I can analysis about work process and according that start create robust python script. Let's discuss about reasonable cost and time for quality work. Note : I will also provide you technical support of this script.

@csscientistsoha6

Hi there, I’m Muhammad Sohaib. I’ve carefully read your project details and I’m confident I can deliver exactly what you’ve described with high quality and on time. Let’s discuss your requirements and get started right away. Best regards, Muhammad Sohaib

@ahmadraza001912

As a seasoned full-stack developer specializing in Python (including the libraries your project seeks on OpenCV and PyMuPDF) and skilled in image processing, I am confident that I have the expertise you need to efficiently deliver your project. My journey into image processing began with a fascination for data extraction, manipulation, and visualization in unique formats such as scanned PDFs. Over the years, I have successfully implemented similar projects with rigorous DPI handling, deskewing/preprocessing needs for digitalized scans – skills that will be invaluable to your work. Another key advantage I bring to the table is my ability to produce clean, reusable, and well-structured code - which aligns with your technical expectations. This will not only ensure that your project is handled professionally but also foster easy maintainability for any future modifications. Through this journey, I will provide regular feedback throughout the development process so that we can optimize the solution within your desired time frame and budget. In conclusion, partnering with me guarantees that not only will you receive working logic - well beyond mere demo scripts - but also a strong commitment to customer satisfaction. I look forward to displaying my expertise by offering you a tailor-made tool that cleanly extracts photos and signatures from your scanned PDFs while meeting every other nuance of your requirement.

@waleedsheeraz06

Hello, I can build a reliable, production-ready Python system to extract photos and signatures from scanned (image-based) PDFs—this is image processing, not OCR, and I have real experience with it. How I’ll handle it: Convert PDFs to high-DPI images (300–400 DPI) Preprocess scans (deskew, noise removal, contrast normalization) Primary ROI-based detection with tolerance for layout shifts Fallback face detection if photo box detection fails Signature extraction via contour/ink analysis Batch processing with consistent output naming Clean, reusable, well-structured Python code Extras I can include: JSON-based configurable ROIs Logging for success/failure Debug images for failed cases Multi-page fallback if page 1 doesn’t contain the targets I work with OpenCV, PyMuPDF/pdf2image, NumPy, and I understand the realities of scanned documents—not demo scripts. Share the samples and expected outputs, and I’ll deliver working logic you can rely on.

@ravichirawa1421

I’m a Python developer with strong OpenCV and image-processing experience. I can extract photos and signatures from scanned PDFs reliably, handle deskewing, DPI variations, and provide batch processing with clean, reusable code.

@Lowk1ee

I understand this is a pure image-processing problem, not OCR. Problem understanding PDFs are fully scanned (image-based) Photo and signature are embedded inside the scanned page Layout is mostly consistent but may shift Photo extraction must fall back to face detection if ROI fails Signature extraction must be ink/contour-based Output filenames must match the source PDF Batch processing is required Proposed solution Convert PDFs to images at 300–400 DPI using PyMuPDF / pdf2image Preprocessing: grayscale, denoise, adaptive thresholding, deskew Photo extraction Primary: configurable ROI (JSON-driven) Fallback: OpenCV Haar/DNN-based face detection Signature extraction Morphological operations + contour detection Filter by aspect ratio, area, and stroke density Support page-1 first, then fallback to additional pages if needed Implementation details Python + OpenCV + NumPy Folder-based batch processing Clean, modular code (reusable functions/classes) Logging for success/failure Debug image dumps for failed cases Config file (JSON) for ROIs and thresholds Why I’m a good fit Hands-on experience with scanned document pipelines Not dependent on OCR Focused on robust logic, not demo scripts Comfortable iterating based on sample PDFs and feedback Delivery Working script within 7 days Tested against provided samples Clear usage instructions I can start immediately once sample PDFs are shared.

@pavitra2004

Hello, I have hands-on experience building OpenCV-based solutions for scanned, image-only PDFs where photo and signature are embedded inside the document image (no text layer). I clearly understand that this is not an OCR problem, but an image-processing one. I can deliver a robust Python pipeline using OpenCV, NumPy, and PyMuPDF/pdf2image with: DPI-aware PDF to image conversion (300–400 DPI) Deskewing and preprocessing for scanned documents ROI-based photo/signature extraction with JSON config Face detection fallback when photo box detection fails Ink/contour-based signature extraction (transparent PNG) Batch folder processing with proper filename mapping Logging, debug outputs, and multi-page fallback The code will be clean, reusable, and production-ready, not a demo script. Happy to iterate using your sample PDFs.

@tharunreddy15

Hi,I can deliver a robust, production-ready Python solution specifically designed for scanned, image-based PDFs, not text-layer documents. My approach will convert PDFs at high DPI (300–400), apply preprocessing (deskewing, noise reduction), and reliably extract the photo and signature even when boxes shift slightly. For photos, I will use ROI-based detection with a fallback to face detection if layout detection fails; for signatures, I will use contour and ink-based image processing to isolate handwritten regions. The script will support batch folder processing, preserve base filenames exactly as required, and be written in clean, reusable Python using OpenCV, NumPy, and PyMuPDF/pdf2image. I can implement a config-driven ROI (JSON), logging, debug image outputs for failures, and multi-page fallback logic if page 1 does not yield results. This will be a practical, working system—not a demo—validated against your sample PDFs with feedback incorporated.

@adityak0462

I will develop a Python solution to extract photos and signatures from scanned, image-based PDFs. Using OpenCV, NumPy, and PyMuPDF/pdf2image, the system will handle layout shifts, apply box and face detection for photos, contour-based detection for signatures, and save outputs using the original PDF filename.

@shashwat16is

Hey there, I have the relevant experience for PyMuPDF/Adaptive padding/ROI estimation that'll be needed for this project. Let's start off with you sending me a sample file over message and I'll verify if my code works for the same. And then we can proceed to work on the same. Thanks, Shashwat

@laraibg1

As a talented and experienced graphic designer, I understand the importance of precision and attention to detail, especially when handling scanned PDFs. My proficiency in using advanced tools, such as Photoshop, will be an asset in extracting your desired photos and signatures seamlessly. While my core expertise lies in graphic design, I have dedicated ample time to expanding my knowledge in Python image processing libraries like OpenCV and NumPy that this project specifically demands. One of the key aspects of your project is the ability to handle scanned documents adeptly. My previous works testify to this capacity as I've handled similar projects requiring detailed extraction from scanned files, rather than relying on OCR methods. I can assure you of a clean and well-structured codebase that allows easy batch processing and DPI handling according to your specific range. Furthermore, beyond mere logo designing or PDF conversions, I'm passionate about creating solutions that add value to clients' work. My commitment to professional communication ensures that feedback during development is incorporated efficiently, guaranteeing a result that exceeds your expectations. Let's work together on this intriguing project and ensure an outcome that combines quality and cost-effectiveness!

@AhmedMushaf

With a passion for learning and a proven track record in Python and JavaScript, I am ideally placed to tackle the challenges your projects presents. Even better, I have extensive experience with image processing and using libraries such as OpenCV, PyMuPDF, and NumPy, all of which are vital components of developing the reliable script/system you need. And although my skills broaden beyond image processing, that precise background is exactly why you should choose me. Understanding the unique nature of scanned documents instead of relying only on OCR is key to overcoming the project's hurdles. More than simply delivering "demo scripts," I will provide a clean, reusable, well-structured code that will meet each of your requirements, including those additional "Nice to Have (Bonus)" ones. But don't just take my word for it; I'll keep you involved throughout the process, implementing your feedback as we go. Post completion I can even provide explanatory documentation so future users can maintain and build upon it themselves. Replies within a few hours

@LiamKJ

I confidently think that I am a great fit for your project as I exceed all your requirements, especially the need for a clean, professional, user-friendly, and seamless Python script that extracts photos and signatures from scanned PDFs with slight layout shifts. I specialize in image processing with OpenCV, PyMuPDF, and NumPy, handling DPI adjustments, deskewing, batch processing, and creating reusable, integrated, automated solutions. While I am new to Freelancer, I have built up the experience to professionally handle all of my customers projects and have completed projects offsite also I'm offering my services at a discount to build up my reputation. I have taken a great deal of interest to your project and would love to discuss more about it.! Regards, Liam Jasson

@vineet370

Hi, I can help with this manual data-entry task and understand the importance of accuracy and consistency when working with public social-media profiles. I will open each profile URL you provide, carefully capture the exact fields requested such as user name, follower count, listed email, and most recent post date—and enter them into your spreadsheet in Google Sheets or Excel, following your formatting guidelines precisely. I work methodically to ensure every entry matches what appears on the page, with correct spelling, date formats, and number formatting so your pivot tables and checks run smoothly. If any information is missing, unclear, or not publicly available, I will leave a clear comment rather than making assumptions. Before delivery, I’ll double-check all rows for completeness and consistency. The final spreadsheet will be clean, well-organized, and ready for your review and spot checks. Regards, Vineet