GetResizer
Home / Guides / Glossary
Speak the same language as your tools

Image editing terms in plain language

Short definitions of the words that decide how an image is stored, resized and shared. Each entry links to the GetResizer tool or guide where the term does real work.

Pixel and megapixel

A pixel is one colored dot in a digital image. An image's dimensions, written as width × height, count pixels along each edge: a 4000 × 3000 photo contains twelve million pixels, or twelve megapixels. More pixels mean more detail and larger files, but a pixel has no fixed physical size until you decide how large to display or print it.

Resolution

Resolution can mean two different things, and mixing them up causes most size mistakes. Pixel dimensions describe how many pixels an image contains. Output resolution describes how densely those pixels are displayed, usually in PPI. A 1200 × 800 image is 1200 × 800 at any setting; what changes is how large it appears. The image size calculator separates the two.

Aspect ratio

The aspect ratio is the proportional relationship between width and height, such as 1:1 for squares or 16:9 for widescreen. Resizing with an unlocked ratio changes the shape; cropping changes the ratio on purpose by removing edges. The resizer locks the ratio by default, and the crop tool offers common presets with an explanation of each.

Alpha channel and transparency

An alpha channel stores how opaque each pixel is, from fully solid to fully transparent. PNG and WebP can store it; JPG cannot. That is why converting a transparent logo to JPG requires choosing a solid background color, as the PNG to JPG and WebP to JPG guides show. Transparent pixels also flatten to white in some exports, which is a format decision, not a bug.

Lossy and lossless compression

Lossy compression permanently discards information the encoder considers hard to notice, which is how JPG and WebP reach small sizes. Lossless compression, used by PNG, records every pixel exactly and stays faithful but usually produces larger files. Neither process improves the source: converting cannot restore detail that earlier compression already removed.

Compression artifacts

Artifacts are visible consequences of lossy encoding: ringing near sharp edges, blocky patches in flat areas and banding in smooth gradients. They appear when quality is set too low or when a file is re-encoded many times. The compression guide shows how to inspect fine hair, text and high-contrast edges at 100% zoom before choosing a quality level.

Quality setting

A quality slider is an encoder control, not a standardized score. 82% in JPG is not the same visual quality or byte count as 82% in WebP, and two browsers can encode the same image differently at the same setting. Judge exported files by their actual bytes and visible detail; the format lab runs that comparison on your own image.

Re-encoding and generation loss

Every lossy save re-compresses the file and can add new artifacts on top of old ones. Editing an exported JPG, saving it, and editing it again compounds the damage. The reliable pattern is to keep the original and export a fresh copy for each destination, which is why GetResizer never requires you to re-upload an exported file.

PPI and DPI

PPI (pixels per inch) describes how tightly pixels are packed on a screen or in a print layout. DPI (dots per inch) describes printer ink dots and belongs to the printing device, not the file. Changing either number relabels the image; it never adds detail. A 1200 × 800 image printed at 300 PPI is four inches wide. The size calculator performs that arithmetic.

Color profile

A color profile, such as sRGB, tells devices how the numeric color values should look. Without a shared profile, the same file can appear duller or more saturated on different screens. Browsers usually assume sRGB, so web exports that keep the default profile behave predictably. Print workflows need a printer-specific profile, which is also why on-screen CMYK values are approximations, as the color picker explains.

EXIF orientation

Many phone cameras store a rotation flag in file metadata instead of rotating the pixels. Software that honors the flag shows the photo upright; software that ignores it shows it sideways. This is why a phone photo can look correct in one viewer and rotated in another. Judge the visible preview before adding a manual rotation, as the rotation guide recommends.

Raster and vector

Raster formats such as JPG, PNG and WebP store a fixed grid of pixels; scaling them up invents new pixels by interpolation. Vector formats such as SVG store shapes and paths that redraw sharply at any size. Wrapping a raster image in an SVG container does not make it a vector, which the PNG to SVG page demonstrates with a downloadable example.

Data URI and embedded images

A data URI encodes a file's bytes directly inside a text document, most often as base64. The raster-in-SVG wrapper produced by PNG to SVG contains exactly that: an image element holding base64-encoded PNG data. You can open the exported file in a text editor and see it.

Container formats

Some extensions describe a container that can hold several kinds of data. HEIC, Apple's photo container, may store image data plus depth maps or bursts, and decoding depends on the browser environment, as the HEIC to JPG guide notes. PDF is a document container whose pages can hold text, vectors or images; rendering a PDF page produces pixels, which is what the PDF to JPG and PDF to PNG tools do locally.

Where these terms meet your files

Every GetResizer tool applies these definitions to real files in your browser: the editor reports actual dimensions and bytes, the lab compares encoders, and the guides show measured examples rather than universal claims. If a term here conflicts with a result you see, trust the measurement and tell us about it.