Batch Image Scanning — How to Analyze 50 Photos at Once
One photo is a question. Fifty photos is a workflow. Batch scanning turns manual image forensics into a production pipeline that runs entirely in your browser.
When One Photo at a Time Isn't Enough
Single-image tools are great for spot checks. Upload a photo, read its EXIF data, run an AI detection scan, move on. But some workflows involve dozens or hundreds of images that all need the same analysis — and doing them one by one is brutal.
Try it free: Batch Scanner — Scan up to 50 images at once for metadata and forensics. Runs in your browser, no signup needed.
An event photographer delivering 200 photos needs to verify that all metadata is consistent and no GPS coordinates are leaking. A journalist fact-checking a photo set needs to confirm every image came from the same camera on the same day. An investigator triaging a seized device needs to quickly sort thousands of images by face count, AI probability, and content classification. A content moderator reviewing user submissions needs to flag problematic uploads before they go live.
These are batch problems. They need a batch solution — one that processes multiple images in parallel, extracts structured data from each, and presents the results in a sortable, filterable, exportable format.
What Gets Extracted From Each Image
The Batch Scanner runs every image through the same multi-layer analysis pipeline that individual Scanly tools provide, consolidated into a single pass:
EXIF metadata. Camera make and model, lens info, shutter speed, aperture, ISO, date and time of capture, GPS coordinates (if present), and software tags showing which app last saved the file. This is the same data you'd see in the EXIF Checker, extracted automatically for every image in the batch.
File properties. Dimensions (width × height), file size, format (JPEG, PNG, WebP), and color space. These help you spot images that don't belong — a 640×480 JPEG in a set of 4000×3000 RAW exports stands out immediately.
Cryptographic hash. A SHA-256 file hash for each image, enabling exact-duplicate detection. If two files in the batch have the same hash, they're bit-identical copies. For near-duplicate detection across visual similarity, the Duplicate Scanner uses perceptual hashing instead.
Face count. The number of detected faces in each image, with confidence scores. Useful for sorting group photos, flagging images that unexpectedly contain people, or finding the shot where someone blinked and turned away.
AI-generated probability. A confidence score indicating how likely the image is to be AI-generated. In a batch of supposedly authentic photographs, a high AI score on any single image is an immediate red flag. See our guide on detecting AI images for more on how this works.
NSFW classification. Content safety scores across multiple categories — safe, suggestive, explicit. The NSFW Scanner runs on every image, letting you filter out problematic content before it reaches an audience.
💡 Did you know?
Batch scanning 50 images locally means 50 photos analyzed — EXIF, faces, AI score, NSFW classification — without a single byte leaving your device. The entire pipeline runs in your browser using JavaScript and WebAssembly.
Sorting, Filtering, and Finding Outliers
Raw data isn't useful until you can sort through it. The batch results table is sortable by every column, which turns a wall of numbers into an investigation tool:
Sort by camera model to verify all photos came from the same device. If 49 images show "Canon EOS R5" and one shows "Samsung Galaxy S24", that image has a different origin — possibly a submission from a different source, a re-download, or a deliberate substitution.
Sort by date to build a timeline. Images should follow a logical sequence. A photo timestamped three days after the rest of the set may have been added later. A missing timestamp (no EXIF date) suggests the image was processed through a tool that strips metadata.
Sort by GPS presence to find privacy risks. If you're preparing photos for publication, any image with embedded GPS coordinates is a potential location leak. Sorting by GPS lets you identify and strip coordinates before delivery — run them through the Privacy Score tool for a full audit.
Sort by AI score to find generated content hiding in a set of real photos. In a submission of 30 product photos, an AI score of 85% on image #17 means it's probably synthetic — worth investigating before publishing.
Sort by face count to find group shots, portraits, or images that unexpectedly contain people. For real estate photography, any image with a face count above zero needs review — reflections in mirrors, people visible through windows, or bystanders in exterior shots.
Need to analyze a full photo set? Upload up to 50 images and get EXIF, faces, AI scores, and NSFW classification in one pass.
Try Batch Scanner →Export for Downstream Tools
Batch scan results can be exported as CSV or JSON for use in spreadsheets, databases, evidence management systems, or custom scripts. Structured export is what separates a scanning tool from a reporting tool.
A CSV export drops directly into Excel or Google Sheets for further analysis — pivot tables by camera model, charts of capture times, conditional formatting to highlight anomalies. A JSON export feeds into automated pipelines — ingesting results into a CMS for metadata-based image tagging, or into an evidence management system with chain-of-custody tracking.
The export includes every data point: file name, dimensions, file size, hash, camera model, date, GPS coordinates, software tag, face count, AI score, and NSFW scores. No data is omitted or summarized — you get the full structured dataset to work with however you need.
Privacy-First Architecture
Batch scanning is where privacy-first design matters most. Processing 50 images through a cloud API means uploading 50 photos to someone else's server. If those images contain faces, that's biometric data crossing the network. If they contain GPS coordinates, that's location data. If they're legal evidence, that's potential chain-of-custody contamination.
Scanly's batch scanner avoids all of this by running entirely client-side. The neural network models for face detection, AI classification, and NSFW scoring are loaded once into your browser. EXIF parsing happens via JavaScript. Hash computation runs locally. The results table renders on your device. At no point does any image data leave your machine.
This isn't just a technical preference — it's a requirement for many professional use cases. Forensic investigators can't upload evidence to third-party servers. Healthcare organizations can't send patient images to unvetted cloud APIs. Journalists protecting sources can't risk photo metadata being logged by an external service.
Real-World Workflows
Journalist verifying a photo set. A source sends 25 photos from a conflict zone. The journalist batch-scans to confirm all images share the same camera model and date range, checks GPS coordinates against the claimed location, and runs AI detection to ensure none are synthetic. One image with a different camera and no EXIF date gets flagged for additional verification.
Photographer auditing before delivery. Before sending 80 wedding photos to a client, the photographer batch-scans to verify no GPS coordinates are embedded (the venue's private address), confirms all images were taken on the same device, and checks that no embarrassing face-detection surprises lurk in reflective surfaces.
Investigator triaging seized media. A forensic analyst receives a hard drive with 500 images. Batch scanning in groups of 50 quickly separates images by camera source, identifies AI-generated content mixed in with real photos, flags NSFW material for priority review, and generates SHA-256 hashes for evidence logging.
Content moderator screening submissions. A UGC platform receives 40 user-submitted product photos. Batch scanning flags three with high NSFW scores and two with high AI probability. These five get routed to human review while the remaining 35 are cleared automatically.
🔍 Pro tip
After batch scanning, use the Duplicate Scanner on flagged subsets to find near-identical versions. Batch gives you the overview; deduplication cleans up the set.
Limitations to Be Aware Of
Browser memory. Processing 50 high-resolution images simultaneously requires significant RAM. Most modern devices handle this without issues, but older machines or tabs with heavy memory usage may slow down. If you hit performance issues, process in smaller batches of 20 to 30.
Very large files. Individual images over 30 megapixels take longer to process through face detection and AI classification because the browser must resize them before inference. EXIF extraction and hashing remain fast regardless of file size.
Formats without EXIF. PNG, WebP, and GIF files rarely contain camera metadata. The batch scanner will still extract file properties, compute hashes, detect faces, and run AI/NSFW classification — but the EXIF columns will be empty. This isn't a limitation of the tool; those formats simply don't carry the data.
Not a replacement for deep analysis. Batch scanning provides breadth — a quick overview of every image. For depth — ELA, JPEG ghost analysis, thumbnail mismatch, steganography — you still need to run individual images through specialized tools. Batch identifies which images deserve that deeper look.
Common Questions
How many images can I scan at once? Up to 50 per session. The practical limit depends on your device's memory. Most modern devices handle 50 without issues. For very large files, work in smaller batches of 20 to 30.
Are my photos uploaded to a server? No. Everything runs in your browser — EXIF parsing, hashing, face detection, AI scoring, NSFW classification. Your images never leave your device.
What data does it extract? EXIF metadata (camera, date, GPS, software), file properties (dimensions, size, format), SHA-256 hash, face count, AI-generated probability, and NSFW scores per category. All results are sortable and exportable.
Can I scan PNG or WebP files? Yes. Face detection, AI scoring, and NSFW classification work on all formats. Non-JPEG formats will have empty EXIF columns because those formats rarely carry camera metadata.
How do I find suspicious images in the results? Sort by the relevant column. High AI scores flag synthetic images. Missing EXIF suggests metadata stripping. Inconsistent camera models reveal mixed sources. The outlier in a consistent set is where to focus.
Breadth First, Depth Second
Batch scanning isn't about replacing detailed forensic analysis — it's about knowing where to point it. Fifty photos in, structured data out. Sort, filter, flag the anomalies, then send those specific images to the specialized tools that can answer the harder questions. The Batch Scanner gives you the map; the individual tools give you the magnifying glass.
