scanlinebased

Scanlinebased refers to a class of rasterization techniques in computer graphics that generate images by processing horizontal lines, or scanlines, from top to bottom. For each scanline, the algorithm determines where it intersects polygons, then fills the corresponding pixel spans. This approach converts geometric primitives into screen pixels in a systematic, line-by-line fashion, often combining depth testing and texture shading during span filling.

A typical scanline-based implementation uses data structures such as an edge table (ET) and an active edge table (AET). The ET organizes polygon edges by their vertical extents; as the scanline advances, edges enter and exit the AET. Each active edge provides an x-intersection that is updated incrementally, and spans are filled between pairs of intersections. This yields efficient, cache-friendly fill operations and supports extensions for anti-aliasing, perspective-correct texture mapping, and shading.

Historically, scanline-based rendering was central to software rasterizers and early hardware pipelines before depth-buffered rasterization became dominant. The method scales with the number of pixels in the output and performs well for scenes with moderate polygon complexity. However, it requires careful handling of overlapping polygons, holes, and transparency, which can complicate sorting and depth considerations.

In modern graphics, scanline concepts underpin the rasterization stage of GPUs, though implementations are highly optimized and parallelized. The approach remains foundational in software renderers and in educational contexts, where it illustrates core ideas of converting geometric primitives into pixel data and the trade-offs involved in efficient image generation.