公开(公告)号：US4933977A

公开(公告)日：1990-06-12

申请号：US247663

申请日：1988-09-22

申请人： Kazuhiko Ohnishi ,  Shoji Miki ,  Yasuo Fujita ,  Hirofumi Kameyama ,  Katsuji Uosaki

发明人： Kazuhiko Ohnishi ,  Shoji Miki ,  Yasuo Fujita ,  Hirofumi Kameyama ,  Katsuji Uosaki

IPC分类号： G06K9/34

CPC分类号： G06K9/344 ,  G06K2209/01

摘要： A method for identifying a plurality of handwritten connected figures includes identifying and prioritizing branches of the connected figures. A first recognition block is formed extending a predetermined distance from one end of the connected figures. Branches having the lowest priority within the first recognition block are erased until a recognizable figure is obtained within the first recognition block. Then, a second recognition block is formed extending the predetermined distance from the recognizable figure obtained in the first recognition block. As with the first recognition block, the branches having the lowest priority within the second recognition block are erased until a recognizable figure is obtained within the second recognition block.

摘要翻译： 用于识别多个手写连接图形的方法包括识别连接图形的分支的优先级。 第一识别块形成为从连接图的一端延伸预定距离。 在第一识别块内具有最低优先级的分支被擦除，直到在第一识别块内获得可识别的数字。 然后，形成从第一识别块中获得的可识别图形延伸预定距离的第二识别块。 与第一识别块一样，在第二识别块内具有最低优先权的分支被擦除，直到在第二识别块内获得可识别的数字。

公开(公告)号：US4987603A

公开(公告)日：1991-01-22

申请号：US247664

申请日：1988-09-22

申请人： Kazuhiko Ohnishi ,  Shoji Miki ,  Yasuo Fujita ,  Hiroki Tsubota ,  Katsuji Uosaki

发明人： Kazuhiko Ohnishi ,  Shoji Miki ,  Yasuo Fujita ,  Hiroki Tsubota ,  Katsuji Uosaki

IPC分类号： G06K9/46 ,  G06K9/68

CPC分类号： G06K9/6857 ,  G06K9/4604 ,  G06K2209/01

摘要： A letter pattern recognition method preprocesses each character pattern to effect characteristic extraction based on elements such as a loop, a straight line and an arc and is adapted to compare the character pattern with a preregistered classification code after the normalization thereof, the character pattern being subjected to balance and stroke checks if necessary as a result of the comparison. Notwithstanding this, if the character is not defined, the character pattern is identified by means of detail classifications such as those of a divergent mode of the arc, and a length ratio of the elements and the like.

摘要翻译： 字母模式识别方法基于诸如循环，直线和圆弧的元素来预处理每个字符模式以实现特征提取，并且适于在其标准化之后将字符模式与预注册分类代码进行比较，字符模式经受 作为比较的结果，如果需要，进行平衡和中风检查。 尽管如此，如果没有定义字符，则通过诸如弧的发散模式的细节分类以及元素的长度比等来识别字符图案。