首页> 外国专利> AUTOMATIC BUTTERFLY SPECIES IDENTIFICATION SYSTEM AND METHOD, AND PORTABLE TERMINAL HAVING AUTOMATIC BUTTERFLY SPECIES IDENTIFICATION FUNCTION USING THE SAME

AUTOMATIC BUTTERFLY SPECIES IDENTIFICATION SYSTEM AND METHOD, AND PORTABLE TERMINAL HAVING AUTOMATIC BUTTERFLY SPECIES IDENTIFICATION FUNCTION USING THE SAME

机译:自动蝶类识别系统和方法,以及便携式终端具有使用相同的自动蝶类识别功能

摘要

The present invention related to a system for automatically classifying butterfly species which extracts branch length similarity (BLS) entropy data based on the multiple feature points of the wing outline of each butterfly species in a reference image of a butterfly whose species classification is identified, inputs the extracted BLS entropy data to a multilayer perceptron neural network, and classifies the outlines of butterfly species by learning the outline of each butterfly species. The system comprises: an outline data extracting unit (110) for extracting the multiple feature points of the wing outline of each butterfly species from a butterfly species reference image; a learning unit (120) for extracting BLS entropy data from the extracted multiple feature points, inputting the extracted BLS entropy data to a multilayer perceptron neural network consisting of an input layer, a hidden layer and an output layer, and obtaining neural network classification data by learning the inputted data; and a database (130) for storing the obtained neural network classification data. [Reference numerals] (100) System for automatically classifying butterfly species;(110) Outline data extracting unit;(120) Learning unit;(130) Database;(140) Classifying unit;(150) Display unit
机译:本发明涉及一种用于自动分类蝴蝶种类的系统,该系统基于识别出种类分类的蝴蝶的参考图像中的每个蝴蝶种类的翼轮廓的多个特征点,提取分支长度相似度(BLS)熵数据,将提取的BLS熵数据导入多层感知器神经网络,并通过学习每种蝴蝶种类的轮廓来对蝴蝶种类的轮廓进行分类。该系统包括:轮廓数据提取单元(110),用于从蝴蝶物种参考图像中提取每种蝴蝶物种的机翼轮廓的多个特征点;以及学习单元(120),用于从提取的多个特征点提取BLS熵数据,将提取的BLS熵数据输入到由输入层,隐藏层和输出层组成的多层感知器神经网络,并获得神经网络分类数据通过学习输入的数据;数据库(130),用于存储所获得的神经网络分类数据。 [附图标记](100)蝴蝶种类自动分类系统;(110)轮廓数据提取单元;(120)学习单元;(130)数据库;(140)分类单元;(150)显示单元

著录项

  • 公开/公告号KR101334858B1

    专利类型

  • 公开/公告日2013-11-29

    原文格式PDF

  • 申请/专利权人 INSTITUTE FOR BASIC SCIENCE;

    申请/专利号KR20120107566

  • 发明设计人 KANG SEUNG HO;LEE SANG HEE;

    申请日2012-09-27

  • 分类号G06T7;G06F19;

  • 国家 KR

  • 入库时间 2022-08-21 15:44:11

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

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