Görsel & Medya — 17 Ücretsiz Online Araç
Toolpile'da ücretsiz tarayıcı tabanlı görsel & medya araçları. Kayıt yok, dosya yükleme yok — her şey cihazınızda yerel olarak çalışır.
Image work splits into three honest jobs: change the bytes (resize, compress, convert), change the look (crop, flip, watermark, border, blur, grayscale), or read information out (EXIF, color extraction, eyedropper). 17 tools here cover all three, all running on your device via the Canvas API — your photos never upload, which matters more than people think for screenshots of work, ID scans, and personal albums.
When to reach for which image tool
Bytes-changing (the bandwidth and quality calls) — Image Resize (downscale before sending), Image Compress (the quality-vs-size knob with sane defaults at 85), Image Format (PNG ↔ JPEG ↔ WebP ↔ AVIF — pick by use-case, not preference; see the bespoke notes on the Image Compress page for the full decision tree), SVG to PNG (rasterise vector for systems that don't accept SVG).
Look-changing (the framing and prep calls) — Image Crop (free or fixed-aspect like 1:1, 16:9, 4:5 for the platform you're posting to), Flip & Rotate, Image Border (solid, polaroid, or blur-fill — see image-border bespoke for the Instagram-square pad-to-fit math), Image Watermark (text overlay), Image Blur / Censor (gaussian blur for redacting faces/license plates before sharing), Image Grayscale (one-click).
Information-reading — EXIF Viewer (camera model, GPS coords, timestamp — useful for either reading metadata OR confirming what's there before you strip it for privacy), Eyedropper (sample any pixel's exact color), Extract Colors (auto-build a 4-6 color palette from the dominant tones in an uploaded photo).
Specific outputs — QR Code (encode any URL/text into a scannable code with size + error-correction options), Favicon (drop in any image, get the right multi-size .ico bundle), Image to Base64 (build the data URI for inlining in HTML/CSS), Code to Image (turn a code snippet into a Twitter-ready PNG with syntax highlighting).
Privacy: image content is the worst class to upload
Photos and screenshots are the highest-leak-risk class of data on the average person's device. A casually-uploaded image can carry: GPS coordinates of where it was taken (EXIF), a face you didn't realise was in the background (Apple's auto-recognise), a screenshot capturing a Slack thread or banking app row, or an ID document being processed for a tedious form. SaaS image tools that promise "we delete after 1 hour" still mean: present in their server's RAM long enough to be logged, included in any cache layer, accessible to ops staff with database access, subpoena-able. None of these are paranoid — they're documented incidents.
Every tool in this category processes the image entirely in your browser via the Canvas API. There is no upload, no cloud round-trip, no "send to our servers for the heavy lifting" — even for the larger operations (compress, convert, resize). For genuinely confidential photos (work screenshots under NDA, medical scans, personal documents, kids' photos), that's the only acceptable processing model in 2026.
Format pickers (the question every image tool secretly answers)
JPEG — photos at quality 80-85. Universal compatibility (every email client, every legacy tool). The wrong choice for screenshots, line art, and anything with sharp edges (lossy artefacts ring around them).
PNG — screenshots, icons, diagrams, anything with text rendered into pixels. Lossless. Big files for photos (4-10× JPEG); appropriate for graphics where you'd see the JPEG artefacts.
WebP — modern web delivery. 25-35% smaller than equivalent-quality JPEG; universal browser support since Safari 14 (2020). Use when you control where it'll be viewed and want bandwidth savings.
AVIF — even smaller than WebP (another 30-50% saving) but slow to encode and not yet universally accepted by every CMS / email tool. Use at build-time with a fallback `
SVG — vector graphics (logos, icons, anything that needs to scale arbitrarily). Tiny file sizes for line art. Convert to PNG only when the destination doesn't accept SVG (some embed contexts, presentation tools, raster-only printers).
Image Format converts between any of these in-browser; Image Compress applies the quality slider; together they cover ~90% of "my image is too big" or "the wrong format" problems.
Things this category deliberately does NOT include
No background removal. Modern background removal needs a trained model (rembg, ClipDrop, Adobe AI). Those require server-side ML or a 50+ MB WASM blob; both contradict the "single tab, no upload, instant load" pattern. Use remove.bg or a desktop tool.
No upscaling / 'enhance'. Real upscaling (Topaz, ESRGAN) needs trained super-resolution models — same constraint as background removal. The honest in-browser version (bicubic/Lanczos) only smooths existing pixels and can't add detail; it would be misleading to ship.
No batch processing. Each tool runs on one image at a time. For "resize 400 photos to 1080p", a desktop tool (ImageMagick CLI, Affinity Publisher batch action) wins on speed and predictability.
No editing-app features (layers, masks, brush tools). Photopea (free, browser-based) and the desktop GIMP/Affinity/Pixelmator tier own that market; this category is utility-grade, not creative-suite.