A benchmark image dataset for industrial tools

Cai Luo; Leijian Yu; Erfu Yang; Huiyu Zhou; Peng Ren

doi:10.1016/j.patrec.2019.05.011

A benchmark image dataset for industrial tools

Cai Luo^*, Leijian Yu, Erfu Yang, Huiyu Zhou, Peng Ren

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

Robots and Artificial Intelligence (AI) play an increasingly important role in manufacture. One of the tasks is to identify tools in the scene so that the tools can be applied to different assembly purposes. In the AI community, many datasets have been generated and deployed to train robots to recognize individual items, however, these datasets are scene-specific and lack generic background. In this paper, we report our dataset contains photos of 8 objects types that would be easily recognized by qualified workers. This is achieved by gathering images of common tools in a typical factory. The ground truth categories of our dataset are manually labeled by experienced workers, which would be worthy evaluation tools for the intelligence industrial systems. The equipment used and the image collection process are discussed, along with the data format. The mean average precisions range from 64.37% to 78.20%, which bring the possibility for future improvement. The dataset is ideal to evaluate and benchmark view-point variant, vision-based control algorithm for industry robots. It is now public available from https://github.com/tools-dataset/Industrial-Tools-Detection-Dataset.

Original language	English
Pages (from-to)	341-348
Number of pages	8
Journal	Pattern Recognition Letters
Volume	125
Early online date	17 May 2019
DOIs	https://doi.org/10.1016/j.patrec.2019.05.011
Publication status	Published - 01 Jul 2019
Externally published	Yes

Bibliographical note

Funding Information:
The work of C. Luo was supported in part by the National Natural Science Foundation of China under Grant 61701541 , in part by the Shandong Provincial Natural Science Foundation , China under Grant ZR2017QF003 , and in part by the Fundamental Research Funds for the Central Universities under Grant 19CX02021A . The work of H. Zhou was supported in part by UK EPSRC under Grants EP/N508664/1 , EP/R007187/1 , EP/N011074/1 , and Royal Society-Newton Advanced Fellowship under Grant NA160342. The work of P. Ren was supported in part by the Shandong Provincial Natural Science Foundation, China under Grant ZR2019MF019. The authors would like to thank the Oil Industry Training Center in China University of Petroleum for the indutrial tools anotation support.

Publisher Copyright:
© 2019 Elsevier B.V.

Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.

Keywords

Benchmark
Image dataset
Industrial tools

ASJC Scopus subject areas

Software
Signal Processing
Computer Vision and Pattern Recognition
Artificial Intelligence

Access to Document

10.1016/j.patrec.2019.05.011Licence: Unspecified

Cite this

@article{ce6c688d831e4b88bee4019a1f66b425,

title = "A benchmark image dataset for industrial tools",

abstract = "Robots and Artificial Intelligence (AI) play an increasingly important role in manufacture. One of the tasks is to identify tools in the scene so that the tools can be applied to different assembly purposes. In the AI community, many datasets have been generated and deployed to train robots to recognize individual items, however, these datasets are scene-specific and lack generic background. In this paper, we report our dataset contains photos of 8 objects types that would be easily recognized by qualified workers. This is achieved by gathering images of common tools in a typical factory. The ground truth categories of our dataset are manually labeled by experienced workers, which would be worthy evaluation tools for the intelligence industrial systems. The equipment used and the image collection process are discussed, along with the data format. The mean average precisions range from 64.37% to 78.20%, which bring the possibility for future improvement. The dataset is ideal to evaluate and benchmark view-point variant, vision-based control algorithm for industry robots. It is now public available from https://github.com/tools-dataset/Industrial-Tools-Detection-Dataset.",

keywords = "Benchmark, Image dataset, Industrial tools",

author = "Cai Luo and Leijian Yu and Erfu Yang and Huiyu Zhou and Peng Ren",

note = "Funding Information: The work of C. Luo was supported in part by the National Natural Science Foundation of China under Grant 61701541 , in part by the Shandong Provincial Natural Science Foundation , China under Grant ZR2017QF003 , and in part by the Fundamental Research Funds for the Central Universities under Grant 19CX02021A . The work of H. Zhou was supported in part by UK EPSRC under Grants EP/N508664/1 , EP/R007187/1 , EP/N011074/1 , and Royal Society-Newton Advanced Fellowship under Grant NA160342. The work of P. Ren was supported in part by the Shandong Provincial Natural Science Foundation, China under Grant ZR2019MF019. The authors would like to thank the Oil Industry Training Center in China University of Petroleum for the indutrial tools anotation support. Publisher Copyright: {\textcopyright} 2019 Elsevier B.V. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.",

year = "2019",

month = jul,

day = "1",

doi = "10.1016/j.patrec.2019.05.011",

language = "English",

volume = "125",

pages = "341--348",

journal = "Pattern Recognition Letters",

issn = "0167-8655",

publisher = "Elsevier",

}

TY - JOUR

T1 - A benchmark image dataset for industrial tools

AU - Luo, Cai

AU - Yu, Leijian

AU - Yang, Erfu

AU - Zhou, Huiyu

AU - Ren, Peng

N1 - Funding Information: The work of C. Luo was supported in part by the National Natural Science Foundation of China under Grant 61701541 , in part by the Shandong Provincial Natural Science Foundation , China under Grant ZR2017QF003 , and in part by the Fundamental Research Funds for the Central Universities under Grant 19CX02021A . The work of H. Zhou was supported in part by UK EPSRC under Grants EP/N508664/1 , EP/R007187/1 , EP/N011074/1 , and Royal Society-Newton Advanced Fellowship under Grant NA160342. The work of P. Ren was supported in part by the Shandong Provincial Natural Science Foundation, China under Grant ZR2019MF019. The authors would like to thank the Oil Industry Training Center in China University of Petroleum for the indutrial tools anotation support. Publisher Copyright: © 2019 Elsevier B.V. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Robots and Artificial Intelligence (AI) play an increasingly important role in manufacture. One of the tasks is to identify tools in the scene so that the tools can be applied to different assembly purposes. In the AI community, many datasets have been generated and deployed to train robots to recognize individual items, however, these datasets are scene-specific and lack generic background. In this paper, we report our dataset contains photos of 8 objects types that would be easily recognized by qualified workers. This is achieved by gathering images of common tools in a typical factory. The ground truth categories of our dataset are manually labeled by experienced workers, which would be worthy evaluation tools for the intelligence industrial systems. The equipment used and the image collection process are discussed, along with the data format. The mean average precisions range from 64.37% to 78.20%, which bring the possibility for future improvement. The dataset is ideal to evaluate and benchmark view-point variant, vision-based control algorithm for industry robots. It is now public available from https://github.com/tools-dataset/Industrial-Tools-Detection-Dataset.

AB - Robots and Artificial Intelligence (AI) play an increasingly important role in manufacture. One of the tasks is to identify tools in the scene so that the tools can be applied to different assembly purposes. In the AI community, many datasets have been generated and deployed to train robots to recognize individual items, however, these datasets are scene-specific and lack generic background. In this paper, we report our dataset contains photos of 8 objects types that would be easily recognized by qualified workers. This is achieved by gathering images of common tools in a typical factory. The ground truth categories of our dataset are manually labeled by experienced workers, which would be worthy evaluation tools for the intelligence industrial systems. The equipment used and the image collection process are discussed, along with the data format. The mean average precisions range from 64.37% to 78.20%, which bring the possibility for future improvement. The dataset is ideal to evaluate and benchmark view-point variant, vision-based control algorithm for industry robots. It is now public available from https://github.com/tools-dataset/Industrial-Tools-Detection-Dataset.

KW - Benchmark

KW - Image dataset

KW - Industrial tools

U2 - 10.1016/j.patrec.2019.05.011

DO - 10.1016/j.patrec.2019.05.011

M3 - Article

AN - SCOPUS:85065915451

VL - 125

SP - 341

EP - 348

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

ER -

A benchmark image dataset for industrial tools

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this