首页> 外文会议>Databases in parallel and distributed systems >Integration of a structural features-based preclassifier and aman-machine interactive classifier for a fast multi-stroke characterrecognition
【24h】

Integration of a structural features-based preclassifier and aman-machine interactive classifier for a fast multi-stroke characterrecognition

机译:集成了基于结构特征的预分类器和人机交互分类器,用于快速多冲程字符承认

获取原文

摘要

A transputer-based parallel machine for handwritten characterrecognition is proposed. An algorithm based on structural features andon a tree classifier was used to accomplish the pre-classification ofthe unknown sample in order to speed up the recognition process. Thealgorithm for the final classification is based on the description ofthe strokes through Fourier descriptors. The learning phase isaccomplished through a man-machine interactive process. The proposedsystem can expand its knowledge base. A special representation of thisknowledge base is proposed in order to record a great amount of data ina suitable way. A fast multistroke handwritten isolated characterrecognition system is presented. The test of this system was performedon a PC based prototype while the realization of a parallel transputerbased working machine is in progress. Experimental results obtainedapplying these machines to handwritten numerals recognition are reported
机译:基于晶片机的手写字符并行机 提出认可。一种基于结构特征的算法 在树分类器上用于完成对 未知样本,以加快识别过程。这 最终分类的算法基于 通过傅立叶描述符描边。学习阶段是 通过人机交互过程来完成。建议 系统可以扩展其知识库。对此的特殊表示 提出了知识库,以便在其中记录大量数据 一种合适的方式。快速的多笔手写孤立字符 介绍了识别系统。对该系统进行了测试 在基于PC的原型上,同时实现并行晶片机 基于工作的机器正在进行中。获得实验结果 报告了将这些机器应用于手写数字识别的情况

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号