用語集

Alpha Channel

Alpha Channel (Transparency Mask)

An alpha channel stores transparency information for each pixel in an image, with values ranging from fully transparent to fully opaque. It is the fourth channel in RGBA color, enabling smooth blending of images over backgrounds.

Aspect Ratio

Aspect Ratio (Width-to-Height Proportion)

Aspect ratio is the proportional relationship between an image's width and height, expressed as two numbers separated by a colon. Common ratios include 16:9 (widescreen), 4:3 (classic), and 1:1 (square).

AVIF

AVIF (AV1 Image File Format)

A next-generation image format based on the AV1 video codec that delivers exceptional compression efficiency, often producing files 50% smaller than JPEG at the same visual quality, with support for HDR and wide color gamut.

Chroma Subsampling

Chroma Subsampling (Color Resolution Reduction)

Chroma subsampling reduces the color resolution of an image while preserving full brightness (luminance) detail. Human eyes are less sensitive to color changes than brightness changes, making this an efficient compression technique.

Chromatic Aberration

A lens defect causing color fringing at edges where different wavelengths of light focus at slightly different points.

Color Grading (Image)

Image Color Grading

Adjusting color balance, contrast, and tone curves to achieve a desired mood or visual style in an image.

Compression Artifact

Compression Artifact (Visual Distortion)

Compression artifacts are visible distortions introduced by lossy compression algorithms. Common artifacts include blockiness (JPEG macroblocks), ringing around sharp edges, color banding in gradients, and mosquito noise.

Dithering (Image)

Image Dithering

A technique that simulates additional colors by arranging limited-palette pixels in patterns that blend visually.

Image Histogram

A graphical representation of pixel intensity distribution in an image, used to assess exposure and contrast.

Image Registration

Aligning two or more images of the same scene taken at different times, viewpoints, or by different sensors.

Metadata

Metadata (Image Descriptive Data)

Image metadata is information embedded within an image file that describes its properties, origin, and content. This includes EXIF camera data, IPTC editorial information, XMP descriptive tags, and ICC color profiles.

Resolution

Resolution (Pixel Dimensions)

The total number of pixels in a digital image, expressed as width times height (e.g., 1920 x 1080). Higher resolution means more detail but also larger file sizes and greater processing demands.

Alpha Compositing

Combining semi-transparent images by blending pixel colors according to their alpha channel values.

Image Pyramid

A multi-scale representation of an image at progressively lower resolutions, used in image analysis and blending.

Rasterization

Rasterization (Vector-to-Pixel Conversion)

Rasterization is the process of converting vector graphics (mathematical shapes) into raster images (pixel grids). This is necessary when displaying SVG, PDF, or font glyphs on pixel-based screens or exporting to bitmap formats.

Color Space

Color Space (Color Model)

A color space is a specific organization of colors that defines the gamut (range) of representable colors. sRGB is the web standard, Adobe RGB covers more greens and cyans for print, and Display P3 serves modern wide-gamut displays.

Lossy Compression

Lossy Compression (Perceptual Data Reduction)

Lossy compression achieves much smaller file sizes by permanently discarding image data that is less perceptible to human vision. JPEG, WebP lossy, and AVIF use this approach to achieve 10:1 to 50:1 compression ratios.

Bit Depth

Bit Depth (Color Depth / Bits Per Pixel)

Bit depth determines the number of possible color values each pixel can represent. Higher bit depths allow more colors and smoother gradients: 8-bit provides 256 values per channel (16.7M colors), while 16-bit provides 65,536.

Vector Graphics

Vector Graphics (Resolution-Independent Graphics)

Vector graphics represent images as mathematical descriptions of geometric shapes, paths, and text rather than pixel grids. They scale to any size without quality loss, making them ideal for logos, icons, and illustrations.

Raster Graphics

Raster Graphics (Bitmap Image)

A type of digital image composed of a rectangular grid of individually colored pixels, where the image's quality and detail are determined by the total number of pixels and color depth.

HDR

HDR (High Dynamic Range)

An imaging technique that captures or represents a wider range of brightness levels than standard images, preserving detail in both the darkest shadows and brightest highlights of a scene.

WebP

WebP (Web Picture Format)

A modern image format developed by Google that provides both lossy and lossless compression with significantly smaller file sizes than JPEG and PNG, while also supporting transparency and animation.

Sprite Sheet

Sprite Sheet (Image Atlas)

A single image file that combines multiple smaller images arranged in a grid, allowing a browser or game engine to display individual graphics by selecting rectangular regions from the larger sheet.

Thumbnail

Thumbnail (Preview Image)

A small, reduced-size version of a larger image that serves as a quick visual preview, commonly used in galleries, file browsers, and search results to help users identify content without loading full-resolution images.

Crop

Crop (Image Region Selection)

The operation of selecting and keeping only a rectangular portion of an image while discarding everything outside the selected area, used to improve composition, remove unwanted elements, or adjust aspect ratio.

Resize

Resize (Image Scaling)

The operation of changing an image's pixel dimensions by scaling it up (enlarging) or down (reducing), using mathematical interpolation to calculate new pixel values at the target resolution.

Image Optimization

Image Optimization (Web Performance)

The process of reducing image file sizes to the smallest possible amount without unacceptable visual quality loss, improving page load speed, reducing bandwidth consumption, and boosting search engine rankings.

Image Interpolation

Image Interpolation (Pixel Resampling)

Image interpolation is the method used to calculate new pixel values when resizing an image. Different algorithms (nearest-neighbor, bilinear, bicubic, Lanczos) trade speed for quality when scaling images up or down.

ICC Profile

ICC Profile (International Color Consortium Profile)

An ICC profile is a standardized data file that characterizes a color device (monitor, printer, scanner) or color space. Embedding ICC profiles in images ensures that colors are reproduced accurately across different devices.

Convolution

Convolution (Image Filter Kernel Operation)

Convolution is the fundamental operation behind image filters like blur, sharpen, and edge detection. A small matrix (kernel) slides across the image, computing weighted sums of neighboring pixels to produce the output.

Canvas

Canvas (Image Working Area)

In image editing, the canvas is the total working area that defines the boundaries of the image. Resizing the canvas adds or removes space around the existing image content without scaling it.

DPI/PPI

DPI/PPI (Dots/Pixels Per Inch)

DPI (dots per inch) and PPI (pixels per inch) measure image resolution density. PPI describes screen display density, while DPI refers to print output resolution. 72 PPI is standard for web, 300 DPI for print.

Gamma

Gamma (Brightness Transfer Function)

Gamma describes the nonlinear relationship between pixel values and displayed brightness. A gamma of 2.2 (the sRGB standard) means that pixel value 128 is not 50% brightness but approximately 21.8%, matching human perception.

Histogram

Histogram (Tonal Distribution Graph)

An image histogram is a graph showing the distribution of pixel brightness values from shadows (left) to highlights (right). Photographers and editors use histograms to evaluate exposure, contrast, and tonal balance.

Noise Reduction (Image)

Image Noise Reduction

Techniques for removing unwanted grain or artifacts from images caused by high ISO or low-light conditions.

Panorama Stitching

Panorama Image Stitching

Combining multiple overlapping photographs into a single wide-angle or 360-degree panoramic image.

Image Segmentation

Partitioning an image into multiple regions or objects, used in computer vision for scene understanding.

Object Detection

A computer vision task that identifies and locates objects within an image using bounding boxes and class labels.

Feature Matching

Image Feature Matching

Finding corresponding keypoints between two images for alignment, stitching, or 3D reconstruction.

Mipmap

Mipmap (Multum In Parvo)

A set of pre-computed, progressively smaller versions of a texture used to improve rendering speed and reduce aliasing.

Texture Atlas

A large image containing many smaller sub-images, used in 3D graphics and games to reduce draw calls.

Morphological Operations

Morphological Image Operations

Image processing operations (erosion, dilation, opening, closing) that modify shapes based on a structuring element.

IPTC

International Press Telecommunications Council Metadata

A metadata standard for news photos including caption, creator, copyright, and keyword information.

Gamut Mapping

Color Gamut Mapping

The process of converting colors from one color space to another while handling out-of-gamut colors gracefully.

Tone Mapping

HDR Tone Mapping

Converting high dynamic range image data to a displayable range while preserving visual detail and contrast.

Vignetting

Image Vignetting

The darkening of image corners relative to the center, caused by lens optics or intentionally applied as a creative effect.

XMP

Extensible Metadata Platform

An Adobe-created XML metadata standard for embedding editing history, keywords, and rights information in files.

Focus Stacking

Combining multiple images shot at different focus distances to produce one image with greater depth of field.