Research on synthesis data generation method for logo recognition
Aiming at the problem of training sample sparse in Logo recognition task under the deep learning framework,a Logo data synthesis algorithm based on contexts was proposed.The algorithm comprehensively utilizes various types of context information to guide the synthesis of Logo images,such as the inte...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
POSTS&TELECOM PRESS Co., LTD
2018-05-01
|
| Series: | 网络与信息安全学报 |
| Subjects: | |
| Online Access: | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2018043 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841530167622631424 |
|---|---|
| author | Yuchao JIANG Lixin JI Chao GAO Shaomei LI |
| author_facet | Yuchao JIANG Lixin JI Chao GAO Shaomei LI |
| author_sort | Yuchao JIANG |
| collection | DOAJ |
| description | Aiming at the problem of training sample sparse in Logo recognition task under the deep learning framework,a Logo data synthesis algorithm based on contexts was proposed.The algorithm comprehensively utilizes various types of context information to guide the synthesis of Logo images,such as the interior of Logo object,the neighborhood of Logo object,the link between Logo object and other objects and the scene where Logo object lives in.The experimental results on the FlickrLogos-32 dataset show that the proposed algorithm can improve the performance of the Logo identification algorithm (mAP increase by 8.5%) without relying on additional manual annotation,verifying the effectiveness of the synthesis algorithm. |
| format | Article |
| id | doaj-art-c65bc8a7585e4fed8c70cd752e6eb677 |
| institution | Kabale University |
| issn | 2096-109X |
| language | English |
| publishDate | 2018-05-01 |
| publisher | POSTS&TELECOM PRESS Co., LTD |
| record_format | Article |
| series | 网络与信息安全学报 |
| spelling | doaj-art-c65bc8a7585e4fed8c70cd752e6eb6772025-01-15T03:12:47ZengPOSTS&TELECOM PRESS Co., LTD网络与信息安全学报2096-109X2018-05-014213159553295Research on synthesis data generation method for logo recognitionYuchao JIANGLixin JIChao GAOShaomei LIAiming at the problem of training sample sparse in Logo recognition task under the deep learning framework,a Logo data synthesis algorithm based on contexts was proposed.The algorithm comprehensively utilizes various types of context information to guide the synthesis of Logo images,such as the interior of Logo object,the neighborhood of Logo object,the link between Logo object and other objects and the scene where Logo object lives in.The experimental results on the FlickrLogos-32 dataset show that the proposed algorithm can improve the performance of the Logo identification algorithm (mAP increase by 8.5%) without relying on additional manual annotation,verifying the effectiveness of the synthesis algorithm.http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2018043Logo recognitiondata synthesiscontextdeep learningdata augmentation |
| spellingShingle | Yuchao JIANG Lixin JI Chao GAO Shaomei LI Research on synthesis data generation method for logo recognition 网络与信息安全学报 Logo recognition data synthesis context deep learning data augmentation |
| title | Research on synthesis data generation method for logo recognition |
| title_full | Research on synthesis data generation method for logo recognition |
| title_fullStr | Research on synthesis data generation method for logo recognition |
| title_full_unstemmed | Research on synthesis data generation method for logo recognition |
| title_short | Research on synthesis data generation method for logo recognition |
| title_sort | research on synthesis data generation method for logo recognition |
| topic | Logo recognition data synthesis context deep learning data augmentation |
| url | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2018043 |
| work_keys_str_mv | AT yuchaojiang researchonsynthesisdatagenerationmethodforlogorecognition AT lixinji researchonsynthesisdatagenerationmethodforlogorecognition AT chaogao researchonsynthesisdatagenerationmethodforlogorecognition AT shaomeili researchonsynthesisdatagenerationmethodforlogorecognition |