ICDL Image Editing (ICDL) key terms

An image built from a rectangular grid of pixels, each holding a colour value. Quality depends on resolution, and enlarging beyond the native size makes individual pixels visible as blocky artefacts.

Artwork described by mathematical paths, curves and fill attributes instead of pixels. It scales to any size without losing sharpness, which suits logos, icons and clean-edged illustrations.

The smallest addressable square of colour in a raster image. Each pixel stores one colour value, and the count across and down gives the image's pixel dimensions.

The amount of detail an image holds, usually stated as pixels per inch on screen or dots per inch in print. More pixels in the same area means finer detail.

The number of bits used to record each pixel's colour. Eight bits per channel gives 256 levels per channel; higher depths hold many more tonal steps for smoother gradients.

An additive colour model that mixes red, green and blue light to build screen colours. Equal maximum values make white, and it is the standard mode for display output.

A subtractive colour model using cyan, magenta, yellow and black inks. Printers combine these on paper, and its gamut is smaller than RGB, so vivid screen colours may shift.

The range of colours a device or colour space can reproduce. Colours outside the gamut cannot be shown exactly and are clipped or approximated during conversion.

A method that permanently discards some image data to shrink file size. JPEG uses it, and repeated saving at strong settings produces visible blocks and halos around edges.

A method that reduces file size while keeping every pixel exactly as it was. PNG and compressed TIFF use it, so files can be reopened and re-edited without cumulative quality loss.

Information stored in the file that describes the picture rather than its pixels, such as camera model, capture date, exposure settings, keywords and copyright. It can be edited or removed.

An extra channel storing transparency for each pixel alongside its colour. It allows areas to be fully or partly see-through and is essential when compositing images.

The light-sensitive chip in a camera or scanner that converts light into electrical signals. Sensor size and pixel count strongly affect image quality and low-light performance.

The true number of samples per inch a scanner captures, set by its sensor and lens. Interpolated resolution adds invented pixels and reveals no extra real detail.

The total light reaching the sensor, set by aperture, shutter speed and ISO. Correct exposure keeps detail in both shadows and highlights.

A camera setting that compensates for the colour of the light so neutral objects look neutral. It removes colour casts from tungsten, daylight or shade.

Magnification produced by physically moving lens elements, so the picture stays sharp. Digital zoom, by contrast, merely enlarges and crops captured pixels.

One million pixels, used to state a sensor's total pixel count. A 12-megapixel image has about twelve million pixels, which limits how large it prints sharply.

A digital copy of what is shown on a monitor, made with a built-in tool or keyboard shortcut. The result is a raster image saved in a chosen format.

The working area of a new document, defined by pixel dimensions, resolution and colour mode. Its size sets the boundaries for all artwork created in the file.

A transparent sheet stacked in a document that holds part of the artwork. Layers can be reordered, hidden, blended or masked so edits stay separate.

A painting tool that applies colour strokes to the active layer, with adjustable size, hardness, opacity and flow. Presets mimic airbrush, pencil or calligraphy effects.

A gradual blend between two or more colours, available in linear, radial and other styles. It is often used to fill backgrounds or fade an image into transparency.

A tool that places editable type on the canvas. Font, size, colour and alignment can be changed later, and type stays live until rasterised for certain effects.

A defined region of the image that limits where edits apply. Marquee, lasso and magic wand variants choose areas by shape, freehand outline or colour similarity.

A command that trims away the outer parts of an image to a chosen rectangle, reducing pixel dimensions and often improving composition. Cropping discards data permanently.

Changing an image's total pixel count by adding interpolated pixels or removing existing ones. Enlarging softens detail; reducing throws information away but keeps files small.

A greyscale channel attached to a layer that hides or reveals parts of it. Black conceals, white reveals and grey gives partial transparency, all without deleting pixels.

A setting that softens a selection edge by spreading the transition over a number of pixels, so pasted, filled or adjusted areas blend into their surroundings.

A retouching tool that copies pixels from a sampled source point and paints them over another area. It hides blemishes or duplicates objects, with an alignment option.

A non-destructive layer applying tonal or colour changes such as levels, curves or hue shifts. It affects layers beneath and can be hidden, masked or deleted later.

A processing effect that alters pixels, such as blur, sharpen, noise or stylised texture. It is destructive unless applied to a smart object.

A lossy raster format widely used for photographs. It supports millions of colours and small files but offers no transparency or layers, and each save loses quality.

A lossless raster format supporting transparency and high colour depth. It suits web graphics, logos and screenshots, though photographic files are much larger than JPEG.

A flexible raster format that can store uncompressed or losslessly compressed images plus layers, alpha channels and metadata. It is common in print and archiving work.

A vector format written in XML that describes shapes, paths and text. It scales cleanly for web and print and can be styled or animated with code.

Writing a file into a chosen format and settings for a specific purpose. Exporting usually flattens layers and may convert colour mode or compress data.

Merging all layers into one background layer to reduce file size and simplify output. It permanently removes layer structure, so an editable copy should be kept.

A file describing how a device or colour space interprets colour, used to convert images predictably between monitor, scanner, printer and other devices.