| 摘要 |
Images are printed by marks formed in pixel arrays by a scanning print head. During each scan marks are made in a pattern that approximates at least portions of many parallel, separated lines -- angled steeply (best at about 3:1 slope, or at least much greater than 1:1) to the scanning axis and shallowly to the print-medium advance. Areas are left unprinted between the angled lines during one or more earlier scans for each image segment, and filled in during one or more later scans. Preferably the marks are made with liquid ink, and the medium heated to hasten drying. Heating causes an end-of-page paper-shrink defect that accentuates positional error components parallel to the print-medium advance; but the lines at a shallow angle to that advance tend to minimize those components -- so the heating and steeply angled lines together promote high throughput while hiding the end-of-page defects. In practice the mark-forming includes placing marks only at pixels where marks are desired for a given image: the angled lines are incomplete where marks are not desired. The angled lines are at a steepest angle possible within design architecture of the scanning print head and print-medium-advance mechanism -- or the steepest such angle consistent with a roughly equal number of marks per pen scan (for desired images in which all pixels are to be marked) and avoidance of other types of defects. The most highly preferred pattern uses corner-to-corner diagonals in a cell three pixels wide and eight tall; this pattern is rotated to obtain two variants, all put down in three passes. For transparent and glossy media, drying is enhanced by a multipass (preferably six-pass) print mode in which the three maximum-diagonal variants are repeated to provide double density, with half the advance distance. <IMAGE> |
