If you work with images daily, sooner or later you will have to face that dreaded moment: resize, format or optimize hundreds or thousands of files at onceDoing it manually with a graphics editor is insane, not to mention a massive waste of time. The good news is that with ImageMagick and a few related tools, you can automate almost the entire process from the terminal.
In this article we'll see how to take advantage of ImageMagick for batch image optimization and conversioncombining it when necessary with other utilities such as mogrify, cwebp o jpegtranYou'll see practical examples in Linux (bash), but many commands are perfectly adaptable to Windows and macOS.
Why do we need to optimize and batch convert images?
Modifying a single photo is very simple: Any viewer or graphics editor allows you to resize and change the format. with two clicks. The problem starts when it's no longer just one or two photos, but tens of thousands spread across folders or huge libraries.
Think of cases like a modeling site with over 300.000 JPEG and PNG imagesMany of these images are between 1 and 15 MB. If a large part of your audience connects with modest connections (for example, users in Germany with slow ADSL), loading these images on a website becomes a nightmare for both the user and the server.
It is also very common to need reduce the size of photos at once to send by email, share via Dropbox/Google Drive, or prepare presentation materials. Doing it manually is tedious and inefficient; the sensible thing to do is automate the task with scripts and console-based tools.
General options for projects with many images
Before delving into ImageMagick, it's important to understand that there are several strategies for handling it. massive media libraries, like those of a very large website:
- Plugins or bulk conversion tools within the CMS (for example, WordPress plugins that convert to WebP). They may work well in medium-sized libraries, but with hundreds of thousands of images they often fall short, time out, or overload the server.
- Optimization at the CDN layer (For example, a Cloudflare paid plan with optimization/Polish). In this case, the CDN serves optimized versions (or WebP) without you having to reconvert all your original files. It's a convenient and scalable option, although you're dependent on the provider and don't reduce the disk space on the origin server.
- System-level batch processing Using ImageMagick, cwebp, and other CLI tools. It requires a bit more initial work (scripts, testing, backups), but it gives you complete control over formats, qualities, paths, and workflows. It's ideal when you want to clean up an old library or prepare resources in a very specific way.
In many large projects, the best approach is to combine both: Optimize and batch convert historical data with ImageMagick and, at the same time, relying on the CDN to polish and serve the lightest possible version according to the browser.
Basic installation of ImageMagick and related tools
The first thing, obviously, is to have it installed ImageMagick and additional utilities whichever you want to use. In most Linux distributions, you can simply use the package manager:
- Debian/Ubuntu (and derivatives):
sudo apt-get install imagemagick
O well
sudo aptitude install imagemagick - openSUSE, Fedora, etc.: installs the package
imagemagickfrom the official repositories or through the graphical tools of your distro.
On Windows and macOS you can download the installer directly from the ImageMagick official websitewhich provides ready-to-use binaries. During installation, be sure to select the option to add the binaries to the System PATH to be able to use magick, convert o mogrify directly from the terminal.
If the official packages are giving you problems (for example, bugs when converting from XCF In Debian/Ubuntu), there are scripts like “ImageMagick Easy Install (IMEI)” that automatically download, compile, and install a newer version from sources.
Some all-in-one local development environments, such as ServBay, already include Pre-installed ImageMagick, cwebp, jpegtran, djpeg, cjpeg and img2webp and added to the PATH, allowing you to use these tools without additional configuration from the integrated terminal.
Key ImageMagick commands for basic use
ImageMagick is not a single program, but a collection of command-line utilities focused on image manipulation. The most typical ones when optimizing and converting are:
- magick / convert: to transform images (change size, format, apply filters, compositing, etc.). In modern versions, the "main" command is
magick;convertIt usually exists as a symbolic link for compatibility. - mogrify: similar to
convertbut designed specifically for batch process and overwrite the original files unless you tell him otherwise. - identify: displays information about an image (format, dimensions, color depth, etc.).
To get quick facts From a specific image, for example a logo in XCF, you can use:
magick identify logo.xcf
If you want a much more detailed report, add -verbose And you'll have all the important metadata broken down. It's useful for making sure that The conversion process maintains or changes what you want.
Change the format and size of individual images
The most basic task is convert from un formato to otroFor example, from JPEG to PNG:
magick convert fotoorigen.jpg fotoresultado.png
ImageMagick decides the output format based on the final file extensionso if you change .jpg by .png, .webp or whatever you want, it's generated directly in that format.
magick convert fotoorigen.jpg -resize 800x600 -quality 90 fotoresultado.jpg
In this example, the image is adjusted to 800 × 600 pixels and it is established JPEG quality at 90%It's a quick way to reduce weight without noticeable loss in the vast majority of web cases.
You can also use width only or height only (for example, -resize 1200x o -resize x800) so that ImageMagick automatically adjusts the other dimension while maintaining the proportions. If you use the sign ! (for example, -resize 800x600!) forces exactly those dimensions, although the image will be distorted if the aspect ratio does not match.
In case of working with GIMP XCF files with multiple layersIf you convert directly to PNG or JPG, a separate file will be generated for each layer. To avoid this and obtain a single, flattened image, add the option -flatten before the output file.
Mogrify vs Convert: When to Use Each One
A very important difference when working in batches is understanding what exactly each command does:
- convert (or magick convert) It reads an input image, applies the transformations, and generates a new output fileThe original image remains intact.
- mogrify Open the file, apply the changes, and unless you use
-formato-path, overwrites the original fileThis saves disk space, but the changes are irreversible if you don't have a backup.
Typical calls are:
convert imagenoriginal.png -parametro valor imagendestino.png
mogrify -parametro valor imagenoriginal.png
E.g., if you run:
convert torres.jpg -negate torresnegadas.jpg
you will still have two different filesthe original and the negative. However, if you do:
mogrify -negate torres.jpg
the file torres.jpg has been modified And there's no backup unless you've created one beforehand. It's ideal for cleaning up a lot of images when you're worried about space, but you have to be careful.
In both cases the The order of the arguments matters. With convert First you specify the source image, then the parameters, and finally the destination. If you reverse the source and destination, the tool will fail because it can't find the file it's supposed to read. mogrify The parameters come before the filenames; if you put the file first and then the options, nothing will be applied.
Creating canvases and texts from scratch
Another key difference is that Only Convert can create images from scratchWith the modifier xc: You can generate a solid color canvas:
convert -size 500x250 xc:red imagenroja.jpg
If you try to do something similar with mogrify Without a previous file, it will fail because There's nothing to modify.The same applies to generating labels or images with text directly:
convert -size 500x250 -gravity Center -background blue -font FreeMono-Bold label:"Linux Center" cartela_Linux_Center.jpg
This command creates a blue background, centers the text, applies a font, and generates a new image. mogrify you would need starting from an existing fileYou can't invent one out of thin air.
Batch processing with mogrify and shell scripts
When we have hundreds of photos in a folder and we want to apply the same conversion automaticallyShell scripts come into play and mogrify.
#!/bin/bash
mkdir -p convertidas
for img in *.jpeg; do
convert "$img" -resize 800x600 -quality 90 "convertidas/$img"
done
This mini-script tours all images with the .jpeg extension In the current folder, apply the resizing and leave the results in converted/You can adapt it to change the format (for example, so that the output is .png o .webp), add watermarks or any other filter that ImageMagick supports.
When you have many images in the same format and you want overwrite them on the site saving space, mogrify It shines especially brightly. A typical example of converting from XCF to PNG, resizing, and sending the result to another folder would be:
mogrify -resize 128x128 -format png -path "$HOME/imgs_convertidas" $HOME/imgs_originales/*.xcf
mogrify for bulk conversion and standard parameters
The order mogrify It allows you to chain together many transformations at once. A fairly complete example for converting all the PDFs in a folder to JPG, resizing them, adjusting quality, and rotating them would be:
mogrify -format jpg -density 300 -quality 80 -resize 1280x1024 -rotate -90 *.pdf
Some of the most commonly used parameters In the context of optimization and conversion, they are:
- -format: output image format (jpg, png, webp, etc.).
- -density: resolution in dpi (150, 300, 600…), useful when converting PDF or other vector formats to bitmap.
- -qualityCompression quality for lossy formats (especially JPEG and WebP). 80 is usually a good balance for the web.
- -resize: final dimensions, being able to specify only width, only height or both.
- -rotate: rotation in degrees (positive clockwise, negative counterclockwise).
Keep in mind that in some distributions (such as Ubuntu/Debian with ImageMagick 7) they have been established security policies that disable certain sensitive formats by default, such as PDF. If when using mogrify You're getting a permissions error with PDFs; you can edit the file. /etc/ImageMagick-7/policy.xml with administrator privileges and change the line:
If when using mogrify about PDFs you see a permissions error, you can edit the file /etc/ImageMagick-7/policy.xml with administrator privileges and change the line:
<policy domain="coder" rights="none" pattern="PDF" />
for something like:
<policy domain="coder" rights="read | write" pattern="PDF" />
security relaxation You should assess whether it's worth it for you or not; from a purist's point of view it opens a small gap, but in many controlled environments it's acceptable.
Conversion and optimization to modern formats (WebP, optimized JPEG)
Besides ImageMagick, there are specific tools that are very useful for optimize size without losing too much qualityespecially when your goal is the web.
Conversion to WebP with cwebp and img2webp
WebP It is a modern format created by Google that offers compression superior to JPEG and PNG In many cases, both with and without loss. The profit cwebp It is used to convert existing images (JPEG, PNG…) to WebP from the command line:
- Convert a JPEG to WebP with 80 quality:
cwebp -q 80 input.jpg -o output.webp - Convert a PNG to WebP while maintaining transparency:
cwebp -q 80 input.png -o output.webp
To batch process (for example, all images that begin with servbay), you can use a small loop in Shell:
for file in servbay*.{jpg,png}; do
cwebp -q 45 "$file" -o "${file%.*}.webp"
done
The syntax ${file%.*} It is a Shell parameter expansion that removes the original extension, and then adds .webp.
If what you want is create an animated WebP (something similar to an animated GIF, but more efficient), you can pull from img2webp:
img2webp -loop 0 -d 100 frame1.png frame2.png frame3.png -o output.webp
This example combines several images into a single animated file, with a 100 ms delay between frames and infinite loop.
Lossless optimization of JPEG with jpegtran, djpeg, and cjpeg
For JPEG there are a number of classic utilities that are very useful when you want to do lossless optimization or very fine control:
- jpegtranIt rotates, flips, reorders, and optimizes JPEGs without recompressing (lossless operations). It also allows you to generate progressive JPEGs and clean up unnecessary metadata.
- djpeg: decodes JPEG to uncompressed intermediate formats (such as PPM), useful in low-level workflows.
- cjpeg: does the opposite process, compressing raw images to JPEG with different quality parameters.
A typical use of jpegtran to optimize JPEGs without loss and making them progressive would be:
jpegtran -optimize -progressive -copy none input.jpg > output.jpg
Here we remove metadata (EXIF, comments, etc.), improve internal organization, and convert to progressive, which usually give better load perception on the web when the image is displayed in "layers" of detail.
Batch image processing with a graphical interface: Converseen
If the terminal isn't your thing, but you still want to take advantage of ImageMagick's power, there are tools like talk, which offers a multiplatform graphical interface (Windows and Linux) for managing bulk conversions.
Converseen is internally based on the ImageMagick library, and allows in one fell swoop:
- Convert between formats (JPG, PNG, TIF, and many others such as DPX, EXR, GIF, SVG, PhotoCD, PostScript…).
- Resize, rotate, and flip batch images.
- Compress images for web or for documents, controlling quality and size.
- Rename file groups with prefixes, suffixes, or progressive numbers.
- Convert a full PDF into many imagesone per page.
Installs easily from the repositories of many distros (converseenand in openSUSE converseen-lang (for the language) or by downloading the executable/portable version on Windows.
The Converseen interface is organized into several panels: action panel (preview, dimensions, rotation, and output options), a central area where you see the files to be processed, and a bottom panel where you choose the output format, specific preferences (JPG/PNG compression, color for transparent background, etc.).
Among its most practical functions are those of massive scale and renownedYou can, for example, select "Scale Image," maintain the aspect ratio, set the width or reduction percentage, and choose whether to keep the original format or convert it to another. For renaming, you have several options. Prefix/Suffix (insert strings around the original name) and Progressive Number (add sequential numbering).
Also allows rotate images 90°, 180° or flip them horizontally/vertically, and it has a specific workflow for converting a PDF to a set of images: you import the PDF, select the pages you're interested in, choose dimensions, format, progressive renaming, and, if necessary, You replace transparent backgrounds with a specific color (for example, white).
Typical practical cases with ImageMagick
With all of the above, you can now set up a wide variety of workflows. Some common examples of Batch optimization and conversion with ImageMagick are:
- Mass resizing For a blog or gallery: use a bash script that iterates through all the large photos, reduces them to a maximum width (e.g., 1200px) and saves them in a "web" folder.
- Download images from a URL, scale and optimize in a single command:
convert https://blogandweb.com/INPUT_IMAGE.jpg -resize 1200x -quality 60 -strip OUTPUT_IMAGE.jpg - Add logos or watermarks automatically apply all images in a campaign using combinations of
-gravity,-geometryy-compositewithin a loop. - remove borders from a set of images with
-shave 10x10, ideal for scanned photos with unwanted frames. - Combine several images into a montage horizontal with
+append, useful for before/after comparisons or frame strips. - Generate animated GIFs from a sequence of JPG/PNG with
-delayy-loop.
All this potential can be orchestrated from other languages such as PHP, Python or Node.js calling the commands using exec(), subprocess o child_processThis makes integrating image optimization into an upload workflow or backend processing quite straightforward.
Master ImageMagick, mogrify and complementary tools such as cwebp or jpegtran It allows you to confidently tackle massive image libraries, reduce website loading times, save disk space, and automate tasks that would be impossible to manage manually. Once you get used to writing three or four well-thought-out commands or a short script, changing the size, format, and quality of thousands of photos ceases to be a chore and becomes a natural part of your workflow. Share this information and more users will learn about the tool.