Presentation | 2015-09-04 Modification of light intensity for image recognition using regression analysis Hiroo Tsuji, Yohei Fukumizu, Takakuni Douseki, Hironori Yamauchi, Ayumi Yoshikawa, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Various image recognition methods are proposed for reading the contents of the license plate in security cameras. In order to improve the image recognition accuracy, a technique for modification of light intensity in the license plate is important. In this paper, we propose a method for modification of light intensity using regression analysis. The proposed method controls the amount of modification by learning the parameters of the regression function from the luminance of the license plate image to adapt to changes in lighting conditions. The image obtained from the proposed method is the subtraction image which preserve image information so that the proposed method is effective in image contrast improvement. In this paper, we compared the proposed method with the Retinex-based image enhancement in respect of visual evaluation of the image quality and image recognition accuracy. As a result, the image contrast of the proposed method was higher than the image contrast of the Retinex-based method. Moreover, the image recognition accuracy of the proposed method was better than the image recognition accuracy of the Retinex-based method, and the difference of the fitness increase ratio between the proposed method and the Retinex-based method is about 11% on average. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | image enhancement / tone correction / luminosity compensation / regression analysis / image recognition / Retinex theory |
Paper # | SIS2015-25 |
Date of Issue | 2015-08-27 (SIS) |
Conference Information | |
Committee | SIS / IPSJ-AVM |
---|---|
Conference Date | 2015/9/3(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Kansai Univ. |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Smart Multimedia Systems, etc. |
Chair | Mitsuji Muneyasu(Kansai Univ.) / Wataru Kameyama(Waseda Univ.) |
Vice Chair | Hirokazu Tanaka(Hiroshima City Univ.) / Takayuki Nakachi(NTT) |
Secretary | Hirokazu Tanaka(Nagoya City Univ.) / Takayuki Nakachi(Toshiba) / (Waseda Univ.) |
Assistant | Hiroyuki Tsuji(Kanagawa Inst. of Tech.) / Hakaru Tamukoh(Kyushu Inst. of Tech.) |
Paper Information | |
Registration To | Technical Committee on Smart Info-Media System / Special Interest Group on Audio Visual and Multimedia Information Processing |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Modification of light intensity for image recognition using regression analysis |
Sub Title (in English) | |
Keyword(1) | image enhancement |
Keyword(2) | tone correction |
Keyword(3) | luminosity compensation |
Keyword(4) | regression analysis |
Keyword(5) | image recognition |
Keyword(6) | Retinex theory |
1st Author's Name | Hiroo Tsuji |
1st Author's Affiliation | Forensic Science Laboratory, Shiga Prefectural Police Headquarters(Shiga Prefectural Police) |
2nd Author's Name | Yohei Fukumizu |
2nd Author's Affiliation | Ritsumeikan University(Ritsumeikan Univ.) |
3rd Author's Name | Takakuni Douseki |
3rd Author's Affiliation | Ritsumeikan University(Ritsumeikan Univ.) |
4th Author's Name | Hironori Yamauchi |
4th Author's Affiliation | Ritsumeikan University(Ritsumeikan Univ.) |
5th Author's Name | Ayumi Yoshikawa |
5th Author's Affiliation | Konan University(Konan Univ.) |
Date | 2015-09-04 |
Paper # | SIS2015-25 |
Volume (vol) | vol.115 |
Number (no) | SIS-208 |
Page | pp.pp.63-68(SIS), |
#Pages | 6 |
Date of Issue | 2015-08-27 (SIS) |