首页> 外国专利> APPARATUS AND METHOD FOR DETERMINING WHETHER TO INCLUDE SPECIFIC ANIMAL CATEGORIES BASED ON DEEP LEARNING

APPARATUS AND METHOD FOR DETERMINING WHETHER TO INCLUDE SPECIFIC ANIMAL CATEGORIES BASED ON DEEP LEARNING

机译:用于确定是否包括基于深度学习的特定动物类别的装置和方法

摘要

The apparatus for confirming the inclusion of an animal category according to an embodiment of the present invention includes a normalization unit for generating a normalized image by normalizing an animal image corresponding to a specific category, and extracting a feature by inputting the normalized image into a pre-learning model, and the feature An initialization unit for initializing the parameters of a triplet network according to, a learning unit for learning the triplet network by inputting a normalized image to the triplet network, a positive distance value and a negative distance by inputting a verification target image to the triplet network And a distance calculation unit that calculates a value, and a category inclusion check unit that generates category-inclusion information by comparing the positive distance value and the negative distance value.
机译:用于确认根据本发明实施例的用于确认动物类别的装置包括归一化单元,用于通过归一化与特定类别对应的动物图像来生成归一化图像,并通过将归一化图像输入预先来提取特征 - 专用模型,并且特征是根据用于通过输入到三联网网络,正距离值和负距离来学习Triplet网络的学习单元,用于初始化三联网网络的参数的初始化单元验证目标图像到三联网网络和计算值的距离计算单元,以及通过比较正距离值和负距离值来生成类别包含信息的类别包含检查单元。

著录项

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