ICDL Image Editing (ICDL) key terms

A picture built from a fixed grid of pixels, each storing a colour value. Enlarging it too far exposes individual squares and softens edges, because existing pixels are stretched rather than newly drawn.

A picture described by mathematical paths, curves and shapes instead of pixels. It stays sharp at any size, which makes it suitable for logos, diagrams, icons and text-based artwork.

The smallest single square unit of a raster image, holding one colour value. The number of pixels across and down a picture sets the maximum detail it can contain.

The amount of detail an image holds, usually expressed as pixel dimensions such as width by height, or as pixels per unit of length when a printed output size is planned.

Pixels or dots per inch: a measure of how densely image data is packed when shown on screen or printed. Raising the value for the same pixel count shrinks the output size.

The number of bits used to record the colour of each pixel. Greater depth allows more distinct colours and smoother transitions; eight bits per channel gives 256 levels per channel.

An additive colour model that mixes red, green and blue light, used by screens, scanners and most digital cameras. All three at full strength create white.

A subtractive colour model using cyan, magenta, yellow and black inks, used for printing. Many bright screen colours fall outside its reach and shift noticeably when converted.

The proportional relationship between an image's width and height, such as 3:2 or 16:9. Keeping it locked prevents stretching or squashing when resizing and cropping.

Information stored inside an image file that describes the picture rather than showing it, such as camera model, capture date, exposure settings, keywords or copyright. It can be viewed, edited or removed.

One million pixels, used to describe a camera sensor and the dimensions of the photos it records. Higher counts allow bigger prints but do not by themselves guarantee better quality.

A camera control that compensates for colour casts caused by different light sources, so neutral objects record as neutral. It may be automatic, chosen from presets or set manually.

The true number of samples a scanner takes per inch, set by its sensor. Interpolated resolution adds invented pixels through software and does not capture extra real detail.

Bringing picture files into an editing application or library from a camera, memory card, scanner or folder, often with options to rename, tag, rotate or convert them on arrival.

An isolated region of the image that edits, filters and paint strokes will affect, while everything outside it stays protected. Selections can be drawn, colour-based or stored for reuse.

A selection created by dragging a rectangular or elliptical outline across the image. Modifier keys typically add to or subtract from the shape, making it useful for regular geometric areas.

A selection tool for tracing irregular outlines, either freehand or by clicking a series of points. A magnetic variant snaps the outline to strong colour edges for difficult shapes.

A selection tool that picks adjoining pixels of similar colour or brightness, judged against a tolerance setting. A higher tolerance captures a wider spread of shades in one click.

Trimming an image down to a smaller area in order to remove distractions or change its proportions. Cropping discards pixels permanently but leaves the remaining pixels unchanged.

Changing how many pixels an image contains. Shrinking throws pixels away and can lose fine detail; enlarging manufactures new pixels by interpolation, which usually softens the result.

A graph counting how many pixels sit at each brightness level. It exposes clipped shadows and blown highlights and helps decide which tonal adjustments an image needs.

Tonal adjustment controls that remap brightness values. Levels sets black, white and midtone points from a histogram; curves does the same with a draggable line graph, giving finer control.

Hue is the colour family itself, such as red or blue; saturation is how intense or grey that colour appears. Adjusting them changes colour character without necessarily altering brightness.

A retouching brush that copies pixels from a sampled source point and paints them elsewhere. It is used to cover dust, blemishes or small unwanted objects with matching surrounding content.

A repair tool that borrows texture from a sampled area but keeps the colour and lighting of the destination, blending the patch in. It suits skin, scratches and small blemishes.

A calculated operation applied to a layer or selection to change its look, such as blur, sharpen, distort or stylise. Some filters stay adjustable afterwards instead of being permanently applied.

Working so the original pixels stay intact and changes are stored as separate adjustments, layers or masks. Any step can then be revised, hidden or removed at a later time.

How much a layer or brush stroke covers what lies beneath it, ranging from fully see-through to fully solid. Lowering it creates ghosting, fading and blended overlays.

One stackable level inside an image document that carries its own content and transparency. Layers let separate elements be edited or moved independently before being merged together.

A greyscale attachment that hides or reveals parts of its layer without deleting pixels. Black conceals, white shows and grey gives partial transparency, and the mask can be repainted later.

A rule deciding how a layer's colours interact with the layers underneath, for example Multiply, Screen or Overlay. Modes produce darkening, lightening, contrast or colour-shift results.

The container and encoding used to store a picture, such as JPEG, PNG or TIFF. Each choice trades off compression, transparency, layer support, colour depth and software compatibility.

Encoding image data more efficiently so the file takes up less space. Lossless methods restore every original pixel; lossy methods discard detail, and repeated lossy saving steadily degrades quality.

A very common lossy format for photographs that gives small files but no transparency and no layers. Its quality slider trades visible artefacts against file size.

A lossless format that supports full transparency and keeps sharp edges, making it a good choice for logos, screenshots and web graphics rather than photographic archives.

A flexible, generally lossless format that can retain high colour depth, layers and print-quality data. It is used for archiving and professional print work, producing large files.

Saving a document out to a chosen file format and settings for a particular destination, such as web, print or another program. Exporting often flattens layers and converts the colour mode.

Choosing format, pixel dimensions, compression level and colour settings so an image downloads quickly and still looks acceptable on screen, where high print resolution is wasted.

A visible or hidden mark placed over an image to assert ownership or discourage unauthorised reuse. It should be applied to an exported copy, never to the master file.