Learning accurate personal protective equipment detection from virtual worlds
- PDF / 1,664,503 Bytes
- 13 Pages / 439.642 x 666.49 pts Page_size
- 98 Downloads / 226 Views
Learning accurate personal protective equipment detection from virtual worlds Marco Di Benedetto1 · Fabio Carrara1 · Enrico Meloni1 · Giuseppe Amato1 · Fabrizio Falchi1 · Claudio Gennaro1 Received: 3 November 2019 / Revised: 4 June 2020 / Accepted: 12 August 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Deep learning has achieved impressive results in many machine learning tasks such as image recognition and computer vision. Its applicability to supervised problems is however constrained by the availability of high-quality training data consisting of large numbers of humans annotated examples (e.g. millions). To overcome this problem, recently, the AI world is increasingly exploiting artificially generated images or video sequences using realistic photo rendering engines such as those used in entertainment applications. In this way, large sets of training images can be easily created to train deep learning algorithms. In this paper, we generated photo-realistic synthetic image sets to train deep learning models to recognize the correct use of personal safety equipment (e.g., worker safety helmets, high visibility vests, ear protection devices) during at-risk work activities. Then, we performed the adaptation of the domain to real-world images using a very small set of real-world images. We demonstrated that training with the synthetic training set generated and the use of the domain adaptation phase is an effective solution for applications where no training set is available.
Marco Di Benedetto
[email protected] Fabio Carrara [email protected] Enrico Meloni [email protected] Giuseppe Amato [email protected] Fabrizio Falchi [email protected] Claudio Gennaro [email protected] 1
Institute of Information Science and Technologies, National Research Council, Pisa, Italy
Multimedia Tools and Applications
Keywords Deep learning · Virtual dataset · Transfer learning · Domain adaptation · Detection · Personal protective equipment
1 Introduction It is estimated that every day around six thousand people die in the world due to accidents at work or occupational diseases, causing more than 2.3 million deaths a year. Many of these accidents could be prevented by the simple use of personal safety equipment, such as helmets or reflective vests. However, it is not always possible to effectively control whether such equipment is actually used. Artificial Intelligence (AI) can be of great help by constantly analyzing the working environment with a camera and warning workers who do not comply with the rules. To this end, the AI sector known as supervised machine learning has achieved significant success in a variety of application domains. These achievements have been so impressive that they have attracted increasing attention from the scientific community to the production of annotated datasets with which to train learning algorithms. In the era of big data, the availability of examples such as images or videos is not considered a
Data Loading...
