Improving Accuracy of Peacock Identification in Deep Learning Model Using Gaussian Mixture Model and Speeded Up Robust Features

Tzu Ting Chen, Ding Chau Wang, Min Xiuang Liu, Chi Luen Fu, Lin Yi Jiang, Gwo Jiun Horng, Kawuu W. Lin, Mao Yuan Pai, Tz Heng Hsu, Yu Chuan Lin, Min Hsiung Hung, C. C. Chen

研究成果: Conference contribution

摘要

Data set is a most crucial aspect for the object recognition by the deep learning. The effect of training a deep learning model would be not good with an insufficient quantity of the data set. For example, there are two similar pictures which are captured from a video with a peacock and these two pictures are separated by one second in this video. Results show that the peacock was recognized by the model in the former picture but it failed in the latter picture due to the angle change of the peacock. In order to improve recognition effects of the model, we propose a system based on Gaussian Mixture Model and Speeded Up Robust Features to recognize peacocks in images. We also implement the prototype of this article and conduct a series of experiments to test the proposed solution. Furthermore, experimental results show the scheme did improve the accuracy of the complete training model.

原文English
主出版物標題Intelligent Information and Database Systems - 12th Asian Conference, ACIIDS 2020, Proceedings
編輯Ngoc Thanh Nguyen, Bogdan Trawinski, Kietikul Jearanaitanakij, Suphamit Chittayasothorn, Ali Selamat
發行者Springer
頁面565-574
頁數10
ISBN(列印)9783030419639
DOIs
出版狀態Published - 2020 一月 1
事件12th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2020 - Phuket, Thailand
持續時間: 2020 三月 232020 三月 26

出版系列

名字Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
12033 LNAI
ISSN(列印)0302-9743
ISSN(電子)1611-3349

Conference

Conference12th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2020
國家Thailand
城市Phuket
期間20-03-2320-03-26

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

指紋 深入研究「Improving Accuracy of Peacock Identification in Deep Learning Model Using Gaussian Mixture Model and Speeded Up Robust Features」主題。共同形成了獨特的指紋。

  • 引用此

    Chen, T. T., Wang, D. C., Liu, M. X., Fu, C. L., Jiang, L. Y., Horng, G. J., Lin, K. W., Pai, M. Y., Hsu, T. H., Lin, Y. C., Hung, M. H., & Chen, C. C. (2020). Improving Accuracy of Peacock Identification in Deep Learning Model Using Gaussian Mixture Model and Speeded Up Robust Features. 於 N. T. Nguyen, B. Trawinski, K. Jearanaitanakij, S. Chittayasothorn, & A. Selamat (編輯), Intelligent Information and Database Systems - 12th Asian Conference, ACIIDS 2020, Proceedings (頁 565-574). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 卷 12033 LNAI). Springer. https://doi.org/10.1007/978-3-030-41964-6_49