• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>2016 5th International Conference on Informatics, Electronics and Vision >Identification of bird species from their singing
【24h】

Identification of bird species from their singing

机译：通过唱歌识别鸟类

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Biodiversity is one of the mysterious phenomenon that we are understanding more and more as time passes by. Aiding the process of understanding nature is an initiative by BIOTOPE society; an open platform to bring together researchers to retrieve information by identifying bird species from calls or singing of different bird species. The main idea behind this work is to identify bird species from their calls or songs. The work presented in this article explores the automated method to identify bird species based on songs and calls of different bird species. This will help in archiving and identifying new species or subspecies. Moreover, it will help behavioral and ecological studies of birds and aid to the surveillance of bird population. In this work we implement a system that will be able to automatically identify bird species from acoustic recordings. Proposed system's ROC performance analysis shows that using Random Decision Tree we can yield 0.96 area under the curve.
机译：随着时间的流逝，生物多样性是我们越来越了解的神秘现象之一。协助了解自然的过程是BIOTOPE社会的一项举措；一个开放的平台，可以使研究人员聚集在一起，通过从不同鸟类的叫声或唱歌中识别鸟类来检索信息。这项工作的主要思想是从鸟类的叫声或鸣叫声中识别鸟类。本文介绍的工作探索了一种基于不同鸟类的歌声和叫声来识别鸟类的自动方法。这将有助于存档和识别新物种或亚种。此外，它将有助于鸟类的行为和生态研究，并有助于监视鸟类数量。在这项工作中，我们实施了一个系统，该系统将能够从录音中自动识别鸟类。所建议的系统的ROC性能分析表明，使用随机决策树可以产生0.96的曲线下面积。

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