| YANG FAN | 1651133: M, 2回目発表 | 数理情報学 池田 和司, 笠原 正治, 久保 孝富 |
|
title: Application of SsVGMM to Medical Data - Classification with Novelty Detection
abstract: There is a huge demand to apply classification in medical analysis. A traditional classifier requires having training samples from each class. However, in reality, it is possible that the testing set may include classes that are not in the training set. This inevitably causes an issue: data from a undefined class will be assigned to a predefined classes. To tackle this, we propose a semi-supervised variational Gaussian mixture model to perform multi-class classification with novelty detection. Comparing to some popular novelty detection meth- ods, we demonstrate that it gets better performance on a thyroid disease data, by generating the distribution of predefined classes and undefined class, without explicitly setting a threshold. language of the presentation: English | ||
| 藤井 博晃 | 1551089: M, 2回目発表 | 数理情報学 池田 和司, 笠原 正治, 久保 孝富 |
|
title: Improvement and performance evaluation of tracking software for multiple animals photographed from UAV
abstract: In recent years, several tracking software have been developed for the purpose of preprocessing behavior analysis of animals and the like. In this study, tracking of Neophocaena phocaenoides is performed using UMATracker which is an individual tracking framework photographed from UAV, software is improved if necessary, and its performance is evaluated. In this time we will deal with the specific behavior seen in Neophocaena phocaenoides to be tracked and the change of coordinates due to the movement of UAV to be shot and report the progress on improvement robustly tracking the target only. language of the presentation: Japanese | ||
| 大明 洋平 | 1561013: D, 中間発表 | 数理情報学 池田 和司, 金谷 重彦, 久保 孝富 |