Passer au contenu principal

Redimensionneur d'image

Image resizing is one of the most common image processing tasks — resizing for web uploads, social media dimensions, email attachments, or print at specific DPI. Our browser-based resizer handles JPG, PNG, WebP, and GIF, with options to lock aspect ratio, set by percentage, or target a specific file size.

Cliquez ou déposez une image

Partager cet outil
File Converters

À propos de Image Resizer

Image resizing is one of the most common image processing tasks — resizing for web uploads, social media dimensions, email attachments, or print at specific DPI. Our browser-based resizer handles JPG, PNG, WebP, and GIF, with options to lock aspect ratio, set by percentage, or target a specific file size.

Comment l'utiliser

  1. Upload an image file.
  2. Set new dimensions by pixels (width × height), percentage of original, or target file size (KB).
  3. Toggle aspect ratio lock to prevent distortion.
  4. Choose output format and quality, then download.

Formule et méthodologie

Aspect ratio = width / height. Scaled height = new width / aspect ratio. Bilinear interpolation: for enlarging, new pixel value = weighted average of 4 nearest source pixels. Bicubic: 16 nearest pixels (sharper, better for photos). For downscaling: Lanczos (sinc-based) algorithm reduces aliasing. File size estimate: width × height × bytes_per_pixel / compression_ratio.

Cas d'usage courants

  • Social media: Facebook (1200×630), Twitter (1600×900), Instagram (1080×1080)
  • Email: reducing attachment size for sending
  • E-commerce: product images at consistent 800×800 for store listings
  • Web performance: serving appropriately sized images (no 4K image for a 400px slot)
  • Print: setting DPI for specific print dimensions (300 DPI for quality print)

Questions fréquentes

For downscaling: Lanczos (or its variant Lanczos3) produces the sharpest results with minimal aliasing — preferred for photos and detailed images. For upscaling: bicubic is standard (Photoshop's "bicubic smoother" is bicubic with extra sharpening). Bilinear is faster but slightly blurrier. For pixel art or icons with sharp edges, nearest-neighbor (no interpolation) preserves crisp edges. Our tool uses Lanczos for downscaling and bicubic for upscaling by default.
Downscaling is generally lossless in terms of perceptible quality — the image simply has fewer pixels but retains sharpness at its new size. Upscaling (making an image larger) does degrade quality because the algorithm must invent pixel information that wasn't there — this produces blurry results beyond ~150% of the original size. For quality enlargement, AI super-resolution tools (waifu2x, Gigapixel AI) are significantly better than standard interpolation.

Outils connexes

Tous les outils →
Image

Base64 to Image Decoder
Image

SVG to PNG Converter
Image

Image Compressor
Image

Image Format Converter
Image

Image to Base64 Encoder

Intégrer cet outil sur votre site

Gratuit pour usage personnel et commercial. Copiez simplement le code ci-dessous.