Presentation | 2015-03-20 A study on construction of pedestrian detectors adaptive to driving environment referring to false detection tendency Yuki SUZUKI, Daisuke DEGUCHI, Ichiro IDE, Hiroshi MURASE, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Recently, safety driving assistance systems using an in-vehicle camera have been studied actively. In particular, since pedestrian detection by image processing is an important task, various methods have been proposed. In real environment, appearance of pedestrians varies largely due to the difference of driving environments such as weather condition, time, driving scene, etc. Therefore, it is difficult to handle all driving scene by a single detector. This report focuses on the false detection tendency according to various driving environments, and constructs pedestrian detectors adaptive to each environment. By applying a conventional pedestrian detector to an in-vehicle camera image, false detections specific to each driving environment are obtained. By referring to these false detections, the proposed method constructs pedestrian detectors adaptive to a driving environment, and they are used for improving the detection accuracy. We conducted experiments using a public dataset to evaluate the proposed method. As a result, accuracy of the proposed method was improved in comparison with a conventional pedestrian detector. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | adaptive learning / false positive tendency / pedestrian detection |
Paper # | BioX2014-67,PRMU2014-187 |
Date of Issue |
Conference Information | |
Committee | PRMU |
---|---|
Conference Date | 2015/3/12(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Pattern Recognition and Media Understanding (PRMU) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | A study on construction of pedestrian detectors adaptive to driving environment referring to false detection tendency |
Sub Title (in English) | |
Keyword(1) | adaptive learning |
Keyword(2) | false positive tendency |
Keyword(3) | pedestrian detection |
1st Author's Name | Yuki SUZUKI |
1st Author's Affiliation | Graduate School of Information Science, Nagoya University() |
2nd Author's Name | Daisuke DEGUCHI |
2nd Author's Affiliation | Information and Communications Headquarters, Nagoya University |
3rd Author's Name | Ichiro IDE |
3rd Author's Affiliation | Graduate School of Information Science, Nagoya University |
4th Author's Name | Hiroshi MURASE |
4th Author's Affiliation | Graduate School of Information Science, Nagoya University |
Date | 2015-03-20 |
Paper # | BioX2014-67,PRMU2014-187 |
Volume (vol) | vol.114 |
Number (no) | 521 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |