Computer Vision System Based for Personal Protective Equipment Detection, by Using Convolutional Neural Network

Zhafran, Faishal; Ningrum, Endah Suryawati; Tamara, Mohamad Nasyir; Kusumawati, Eny

doi:10.1109/elecsym.2019.8901664

Cited by 10 publications

(5 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This statement is confirmed by the experimental results presented in [21]. Our method is based on a two-stage approach, hence it is substantially different from methods performing detection in a single stage [9,12,28,30,31,32]. Some recent works [3,6] focused on helmet detection of bike or motor riders.…”

Section: Related Worksupporting

confidence: 62%

“…The PPE detection approaches based on deep learning are divided into two categories: (i) training an object detector for PPE item detection or (ii) training a person detector and a classifier that takes the bounding boxes predicted by the person detector as input. The majority of deep learning systems [9,12,28,30,31,32] perform PPE detection in a single step, by training an object detector for the corresponding object classes. For example, Fang et al [9] trained a Faster R-CNN object detector [24] on more than 100,000 images to detect people not wearing helmets.…”

Section: Related Workmentioning

confidence: 99%

“…For example, Fang et al [9] trained a Faster R-CNN object detector [24] on more than 100,000 images to detect people not wearing helmets. Another approach for helmet detection based on Faster R-CNN is presented in [32]. In other works [12,31], the authors approached the task of helmet detection using the YOLO object detector [22].…”

Section: Related Workmentioning

confidence: 99%

“…Methods that detect PPE items in a single stage [9,12,28,30,31,32] are typically less accurate than two-stage methods performing people detection and PPE recognition [3,6,21]. The main issue with methods based solely on object detectors is the inability of object detectors to detect small objects, e.g.…”

Section: Related Workmentioning

confidence: 99%

“…Typical deep learning approaches for PPE detection are (i) to train an object detector [9,12,28,30,31,32], resulting in a single-stage pipeline, or (ii) to train a person detector and a classifier that takes the bounding boxes predicted by the detector and discriminates between people wearing and people not wearing the corresponding PPE items [3,6,21], resulting in a two-stage pipeline. A recent work [21] demonstrated that the latter methodology produces more accurate results, mainly due to the poor ability of object detectors of detecting small objects.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

SuPEr-SAM: Using the Supervision Signal from a Pose Estimator to Train a Spatial Attention Module for Personal Protective Equipment Recognition

Sandru¹,

Duta²,

Ionescu³

2020

Preprint

View full text Add to dashboard Cite

We propose a deep learning method to automatically detect personal protective equipment (PPE), such as helmets, surgical masks, reflective vests, boots and so on, in images of people. Typical approaches for PPE detection based on deep learning are (i) to train an object detector for items such as those listed above or (ii) to train a person detector and a classifier that takes the bounding boxes predicted by the detector and discriminates between people wearing and people not wearing the corresponding PPE items. We propose a novel and accurate approach that uses three components: a person detector, a body pose estimator and a classifier. Our novelty consists in using the pose estimator only at training time, to improve the prediction performance of the classifier. We modify the neural architecture of the classifier by adding a spatial attention mechanism, which is trained using supervision signal from the pose estimator. In this way, the classifier learns to focus on PPE items, using knowledge from the pose estimator with almost no computational overhead during inference.

show abstract

Section: Related Worksupporting

confidence: 62%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

SuPEr-SAM: Using the Supervision Signal from a Pose Estimator to Train a Spatial Attention Module for Personal Protective Equipment Recognition

Sandru¹,

Duta²,

Ionescu³

2020

Preprint

View full text Add to dashboard Cite

show abstract

Advances in the Use of Artificial Intelligence and Sensor Technologies for Managing Industrial Workplace Safety

Vukićević

Petrović

2023

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

The compliance of head-mounted industrial PPE by using deep learning object detectors

Isailović

Peulić

Djapan

et al. 2022

Sci Rep

View full text Add to dashboard Cite

The compliance of industrial personal protective equipment (PPE) still represents a challenging problem considering size of industrial halls and number of employees that operate within them. Since there is a high variability of PPE types/designs that could be used for protecting various body parts and physiological functions, this study was focused on assessing the use of computer vision algorithms to automate the compliance of head-mounted PPE. As a solution, we propose a pipeline that couples the head ROI estimation with the PPE detection. Compared to alternative approaches, it excludes false positive cases while it largely speeds up data collection and labeling. A comprehensive dataset was created by merging public datasets PictorPPE and Roboflow with author’s collected images, containing twelve different types of PPE was used for the development and assessment of three deep learning architectures (Faster R-CNN, MobileNetV2-SSD and YOLOv5)—which in literature were studied only separately. The obtained results indicated that various deep learning architectures reached different performances for the compliance of various PPE types—while the YOLOv5 slightly outperformed considered alternatives (precision 0.920 ± 0.147, and recall 0.611 ± 0.287). It is concluded that further studies on the topic should invest more effort into assessing various deep learning architectures in order to objectively find the optimal ones for the compliance of a particular PPE type. Considering the present technological and data privacy barriers, the proposed solution may be applicable for the PPE compliance at certain checkpoints where employees can confirm their identity.

show abstract

Computer Vision System Based for Personal Protective Equipment Detection, by Using Convolutional Neural Network

Cited by 10 publications

References 2 publications

SuPEr-SAM: Using the Supervision Signal from a Pose Estimator to Train a Spatial Attention Module for Personal Protective Equipment Recognition

SuPEr-SAM: Using the Supervision Signal from a Pose Estimator to Train a Spatial Attention Module for Personal Protective Equipment Recognition

Advances in the Use of Artificial Intelligence and Sensor Technologies for Managing Industrial Workplace Safety

The compliance of head-mounted industrial PPE by using deep learning object detectors

Contact Info

Product

Resources

About