Applying of Adaptive Threshold Non-maximum Suppression to Pneumonia Detection

Hao Teng, Huijuan Lu*, Minchao Ye, Ke Yan, Zhigang Gao, Qun Jin

*この研究の対応する著者

研究成果: Conference contribution

抄録

Hyper-parameters in deep learning are sensitive to prediction results. Non-maximum suppression (NMS) is an indispensable method for the object detection pipelines. NMS uses a pre-defined threshold algorithm to suppress the bounding boxes while their overlaps are not significant. We found that the pre-defined threshold is a hyper-parameter determined by empirical knowledge. We propose an adaptive threshold NMS that uses different thresholds to suppress the bounding boxes whose overlaps are not significant. The proposed adaptive threshold NMS algorithm provides improvements on Faster R-CNN with the AP metric on pneumonia dataset. Furthermore, we intend to propose more methods to optimize the hyper-parameters.

本文言語English
ホスト出版物のタイトルBio-inspired Computing
ホスト出版物のサブタイトルTheories and Applications - 14th International Conference, BIC-TA 2019, Revised Selected Papers
編集者Linqiang Pan, Jing Liang, Boyang Qu
出版社Springer
ページ518-528
ページ数11
ISBN(印刷版)9789811534140
DOI
出版ステータスPublished - 2020
イベント14th International Conference on Bio-inspired Computing: Theories and Applications, BIC-TA 2019 - Zhengzhou, China
継続期間: 2019 11月 222019 11月 25

出版物シリーズ

名前Communications in Computer and Information Science
1160 CCIS
ISSN(印刷版)1865-0929
ISSN(電子版)1865-0937

Conference

Conference14th International Conference on Bio-inspired Computing: Theories and Applications, BIC-TA 2019
国/地域China
CityZhengzhou
Period19/11/2219/11/25

ASJC Scopus subject areas

  • コンピュータ サイエンス(全般)
  • 数学 (全般)

フィンガープリント

「Applying of Adaptive Threshold Non-maximum Suppression to Pneumonia Detection」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル