Abstract: A restricted region transform method and a restricted region transform device are disclosed. The method includes following steps: reading out bare board information of a printed circuit board and layout information of a plurality of components, wherein the layout information of the plurality of components corresponds to a plurality of physical restricted regions; setting a first region according to edge information in the bare board information; setting a plurality of second regions according to projections of the plurality of physical restricted regions on a surface of the printed circuit board; selecting every two second regions, which overlap each other, among the plurality of second regions, and the selected second regions constituting a restriction conflict set; and selectively amending the second regions in the restriction conflict set to remove one or more overlaps from the second regions in the restriction conflict set.

Abstract: A restricted region transform method and a restricted region transform device are disclosed. The method includes following steps: reading out bare board information of a printed circuit board and layout information of a plurality of components, wherein the layout information of the plurality of components corresponds to a plurality of physical restricted regions; setting a first region according to edge information in the bare board information; setting a plurality of second regions according to projections of the plurality of physical restricted regions on a surface of the printed circuit board; selecting every two second regions, which overlap each other, among the plurality of second regions, and the selected second regions constituting a restriction conflict set; and selectively amending the second regions in the restriction conflict set to remove one or more overlaps from the second regions in the restriction conflict set.

Abstract: In speech recognition, an unknown input signal is analyzed for its feature pattern sequence A.sub.i which is compared to prestored reference patterns grouped in categories. Each category m has a representative reference pattern B.sup.m which represents the average value of the n patterns in the catagory B.sub.n.sup.m. A first comparison (prematching) to the representative patterns is followed by a second matching to the actual reference patterns in the most likely category. The representative pattern is used in pre-matching and matching, thereby simplifying calculation and increasing accuracy.