Image Extraction and Folder Organization from Auto Parts PDFs (PDF to Folders)

@Imranatmoon

Hi, I see there is a web version of the pdf in the catalog form on the manufacturer website and I feel its easy to scrape images with their part number to local drive than capture images from PDF file. if you're okay with web version please let me know. I'm available on chat to discuss and ready to start Regards Imran

@jawadaziz78

"Auto Electrical Parts" Hi, I'm Jawad. I do PDF extraction and data-organization work regularly, with a 5-star review from a client in Spain for careful document work. One question: is the Part Number always in a consistent spot next to each photo, or does the layout change between catalogs? Timeline: send one sample PDF and I'll return that catalog's folders within 24 hours so you can confirm accuracy, then the full batch in 3–4 days. Budget: Adjustable once I see the total page count. Reply and I'll start on your sample today. Best Regards, Jawad Aziz

@persisusanm

Hi, manually cropping images across multiple catalogs works, but it's slow and error-prone once you're doing this repeatedly across brands. I'd script this instead: extract embedded images from each PDF (PyMuPDF), OCR each part number label near its photo (Tesseract), then auto-create the folder and save the matched image - accurate, fast, and repeatable if you send more catalogs later. I'll spot-check accuracy manually before final delivery, since correct photo-to-part-number matching is the one thing that can't have errors. Estimated time: 1-2 days depending on catalog count and consistency of layout across PDFs. Ready to start once I have a sample PDF or two to confirm the layout pattern. Arun,susan

@LudmilaSvit

Good afternoon! I'd be happy to help with this project. Before confirming the timeline, could you please clarify: • Approximately how many PDF catalogs are included? • About how many part images need to be extracted in total? • Will the folder structure already exist, or should it also be created from scratch? I noticed the sample images include the dashed borders from the catalog. Should the extracted images keep those borders, or would you like clean crops without them? Once I know the total scope, I can provide an accurate estimate.

@BrocoDevServices

I checked your attached examples: ASC-REG-404A and ASC-REG-405 show the two failure points that matter here—suffixes must stay exact, and flattened catalog pages make simple embedded-image extraction unsafe. I will map each crop to its printed part number, then manually cross-check the image/code pairing. You receive one exact-name folder per part, the source-resolution crop, a CSV manifest (catalog, page, part number, output path), and an exceptions list for unreadable or ambiguous rows. I will preserve source quality; no invented enhancement or renamed codes. Relevant public PDF-QA sample: https://www.freelancer.com/portfolio-items/11492391-pdf-and-document-transformation-qa — an independent service sample, not client work. This ₹900 / 2-day bid covers a paid first batch of up to 150 part images from up to 6 supplied catalogs, plus one consolidated correction round. If the full set is within that cap, it is the full delivery. Before award acceptance, please provide 1–2 representative PDFs, total page/part count, preferred PNG or JPG, and whether borders should be excluded. I will first confirm that the sample supports reliable label-to-image mapping. Work starts only after that check and the full ₹900 milestone is funded. One quick question: roughly how many part images are in the first catalog? If the first batch stays inside the stated cap, the fixed step is ready.