Journal of Physics: Conference Series

PAPER • OPEN ACCESS You may also like

- Defect detection of injection-molded parts
Research on Safety Helmet Detection Algorithm of based on improved-YOLOv5
Haoming Liang, Jianrui Chen, Wei Xie et
Power Workers Based on Improved YOLOv5 al.

- Construction Site Safety Helmet Wearing

Detection Method based on Improved
To cite this article: Desu Fu et al 2022 J. Phys.: Conf. Ser. 2171 012006 YOLOv5
Liang Fu

- A real-time method for detecting bottom

defects of lithium batteries based on an
improved YOLOv5 model
View the article online for updates and enhancements. Yu Zhang, Shuangbao Shu, Xianli Lang et
al.

This content was downloaded from IP address 115.111.246.26 on 11/10/2023 at 10:01

ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

Research on Safety Helmet Detection Algorithm of Power

Workers Based on Improved YOLOv5

Desu Fu1, Lin Gao1*, Tao Hu1, Shukun Wang1, Wei Liu1
1
School of Information Engineering, Hubei MinZu University, Enshi, Hubei,China
*
email: [email protected]

Abstract. The traditional helmet detection algorithm in power industry has low precision and
poor robustness. In response to this problem, the helmet detection algorithm based on
improved YOLOv5 (You only look once) is put forward in this paper. Firstly, the YOLOv5
network structure is improved. By increasing the size of the feature map, one scale is added to
the original three scales, and the added 160*160 feature map can be used for the detection of
small targets; Secondly, the K-means is used for re-clustering the helmet data set to get more
suitable priori anchor boxes. The experimental results illustrate that the average accuracy of the
improved YOLOv5 algorithm is increased by 2.9% and reaching 95% compared with the
initial model, and the accuracy of helmet recognition is increased by 2.4% and reaching 94.6%.
This algorithm reduces the rates of missing detection and misdetection of small target detection
in original network, and has strong practicability and advanced nature. It can satisfy the
requirements of real-time detection and has a certain role in promoting the safety of power
industry.

1. Introduction
In the working process of power workers, our first impression of them is that they are wearing safety
helmets, whether it is in sunny, rainy or snowy. If the power workers do not wear safety helmets
during operation, they may be hit by objects falling from above, hurt their head due to falling from a
height, or their heads may suffer from electric shock. Therefore, safety helmet is the safety guarantee
for workers in power industry.
Power workers must wear safety helmets to enter the operation area, but manual supervision is time
consuming and laborious, and there are risks in close range supervision in some work scenarios. So the
intelligent real-time safety helmet detection system of power workers is particularly important. It can
not only realize the automation and digitization of safety supervision and monitoring, but also improve
the safety of power workers, which has practical development significance.
The development of target detection technology is divided into two periods, which can be called
the traditional detection period and the deep learning-based detection period [1]. In the traditional
target detection period, VJ (Viola-Jones) face detector [2], HOG + SVM (Histogram of oriented
gradient + Support Vector Machine) algorithm [3] and DPM (Deformable Part Model) algorithm [4]
are the representations. For example, in 2014, Liu Xiaohui [5] combined SVM and skin color
detection to Identify helmets. The detection period based on deep learning is represented by R-CNN
(Region-Convolutional Neural Networks) [6], Fast R-CNN [7], Faster R-CNN [8], SPP-Net (Spatial
Pyramid Pooling-Net)[9], YOLO [10], SSD (Single Shot MultiBox Detector)[11] algorithm, etc.
These algorithms are divided into two-stage and one-stage. The former mainly includes R-CNN,
SPP-Net, Fast R-CNN and Faster R-CNN, and the latter mainly includes YOLO, SSD, etc. In the
two-stage algorithm, the first-level network extracts features from the candidate area, and the
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

second-level network classifies and accurately regresses the selected area. The detection accuracy is
high, but the running speed is slow. In the single-stage algorithm, the tasks of classification and
regression can be completed only by the first-level network, and the step of candidate regions is not
required. The running speed is fast and the detection accuracy is slightly lower. The YOLO is a
representative of single-stage algorithm. Because of its fast running speed, it is suitable for real-time
detection. Very popular in practice. The YOLOv5 is the latest and best performance version, and its
application and research are very important.

2. Principle of YOLOv5 algorithm

2.1. Brief description of the development of YOLOv5

The YOLO series have become the current hot target detection algorithms. Compared with other
algorithms, they have the characteristics of fastness and real-time, and their structure are relatively
simple. The method is to first extract features, The input image is then divided into s * s grids, and
finally detect the target whose center point falls on the grid.
After YOLOv1[10] appeared in 2015, in order to continue to improve its performance,
YOLOv2[12], YOLOv3[13], YOLOv4[14] appeared, and they have been updated to the YOLOv5[15].
Take the comparison between YOLOv5 and YOLOv4, the former is more flexible and quicker than
the latter without reducing its accuracy. It includes four models: YOLOv5s, YOLOv5m, YOLOv5l,
and YOLOv5x. Their parameter size and accuracy increase in order,From Bottleneck to distinguish,
there are some mechanisms like EfficienctNet [16] to select a model of the appropriate size.
The version in YOLOv5 is updated very quickly, mainly including the 3.0, 4.0 and 5.0 versions.
The comparison of the models trained by YOLOv5s in the 3.0, 4.0, and 5.0 versions is shown in Table
1. The 5.0 version is not much updated compared to the 4.0 version, and the main update is The
version 5.0 can directly test online videos. Compared with version 3.0, version 4.0 has updated the
new activation function SiLU to replace the original LeakyReLU and Hardswish, as shown in Figure 1
below. This SiLU was also introduced in Pytorch 1.7, the model is more streamlined, and a
convolutional layer is removed in each bottleneck, as well as the reconstruction of the utils module
and so on. This article uses version 5.0 of the YOLOv5s model.
Tab. 1 Performance comparison of models trained by YOLOv5s in version 3.0, 4.0 and 5.0
YOLOv5s size mAPval mAPtest mAPval Spend params FLOPS
(pixels) 0.5:0.95 0.5:0.95 0.5 v100(ms) (M) 640(B)
3.0 640 37.0 37.0 55.4 2.4 7.5 13.2
4.0 640 36.8 36.8 63.1 2.2 7.3 17
5.0 640 36.7 36.7 66.9 2.0 7.3 17

2.2. Model network architecture of YOLOv5

The network structure of YOLOv5 includes four parts: Input, Backbone, Neck, and Prediction. The
structure of YOLOv5 is shown in Figure 1.

2
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

Fig. 1 Network structure of YOLOv5

Firstly, mosaic data enhancement method is used for its capture in the Input terminal, and it also
has an automatic anchor frame mechanism system, which is different from the separate anchor frame
mechanism of YOLOv3 and YOLOv4.
Secondly, the Backbone is divided into two structural domains. One is the Focus structure, which is
unique in YOLOv5. Its slicing operation is a very critical step.For example, the image input of 640 *
640 * 3, after the slice operation, it will turn into a characteristic map of 320*320*12, and then pass
through 32 convolution kernels, eventually become 320*320*32 Feature map. The other is the CSP
(Cross Stage Partial network) structure. YOLOv5 has designed two CSP structures, the CSP structure
is applied to the backbone and the neck network. The SPP network module is also used in the
Backbone part. The SPP network was proposed by He Kaiming [9] in 2015. The purpose of using SPP
is that regardless of the input size, it can generate a fixed size output, and it can also use multiple
pooling windows.
Thirdly, the Neck part of the current version of YOLOv5 adopts the structure of FPN+PAN
(Feature Pyramid Networks + Path Aggregation Networks)[17]. The FPN transmits semantic
information from high dimension to low dimension (the big goal is more clear), and the PAN transmits
semantic information again from low dimension to high dimension (the small goal is also more clear).
In the neck, the CSP structure designed by CSPNet[18] is also adopted and used for reference, It

3
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

enhances the ability of network feature fusion. In addition, other parts of the network have also been
adjusted.
Finally, the Prediction of YOLOv5 is innovative to a certain extent. It Increased PANNet [19] to
better complement the underlying and high-rise feature advantages, effectively solve the multi-scale
problem.

3. Improved YOLOv5 helmet detection algorithm

The YOLOv5 is a real-time, fast and accurate algorithm. However, for the detection of some small
targets, YOLOv5 may have false detection or missed detection. In addition, due to the operation of
convolution and down sampling, the number of feature maps will be reduced, and it is easy to lose
feature information in the transmission process, so it is easy to produce gradient disappearance. For
the sake of improve the detection accuracy and effect, a safety helmet detection algorithm for power
workers according to improved YOLOv5 is put forward. The main improvements are as follows:
 Improve and optimize the feature fusion layer and multi-scale detection layer, and add a
fusion scale layer for small target detection, which greatly improves the detection ability of
small targets or even dense small targets.
 Because there are only two categories for identification and detection of helmets, and the
priori anchor frame obtained in the original YOLOv5 algorithm clustering by K-means for
COCO data set is not suitable for the actual recognition and detection of helmets, so K-means
is accustomed to re-cluster the priori anchor frame that is more suitable for detecting and
identifying helmets.
 Based on the experimental data, this paper selects between GIOU Loss and CIOU Loss, and
finally selects the CIOU Loss with better effect as the loss function of bounding box
regression. Because CIOU_ Loss also considers the scale information of the width height ratio
of the bounding box.

3.1. Improvement of feature extraction network

In the YOLOv5 algorithm, its feature fusion layer and detection layer use the FPN and PAN network
structure to enhance the ability and increase the accuracy of image recognition. The minimum feature
scale layer size output is 20*20, and the smaller feature map contains a lot of semantic information,
but the error will be large for the information judgment of the predicted position.The detection of the
helmets may be affected by different positions, distances, weather conditions, and location occlusion.
Small targets such as safety helmets are prone to misdetection or missed detection. The purpose is to
better improve the detection ability of small targets, we have added a scale layer for small targets on
the original three output scales. After fusion with other feature maps, four output scale layers are used
to identify and detect safety helmets.
The original three scale output detection layers are 20*20, 40*40 and 80*80, which are used to
detect large, medium and small targets respectively. If the detection of safety helmets belongs to small
targets, this paper makes the following changes on the original basis: first, because the detection is
small targets, in the case of the original three priori frames,a priori box for small target detection is
added, and the network of the initial YOLOv5 model is changed. The difference between the
improved YOLOv5 network model and the original algorithm is that if the size of the input picture is
640*640*3, after the second up sampling splicing in the neck part, another up sampling will be carried
out through C3 module and CBS module (The two modules are shown in Figure 1), It is spliced with
the 160*160*64 features output in the backbone part, which becomes the size of 160*160*128 after
the operation of C3 module. Through the up sampling operation in the feature fusion layer stage, the
image is finally output to the size of 160*160*255.
Therefore, compared with the output of the original YOLOv5 network model, the improved
network model increases a 160*160 output layer. That is, the improved output layer size is 160*160,
80*80, 40*40, 20*20. The feature map size of 160*160 is 4 times downsampling of the input image
640*640, the receptive field is relatively small, suitable for detection and identification of relatively
small targets such as helmets. The helmet detection model trained on the above 4 scale output layers

4
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

can be accurately detected on multiple scales even if the size of the helmets in the screen even have
some changes. The improved YOLOv5 model is shown in Figure 2.

Fig. 2 The network structure of our improved YOLOv5(The red box is the improved part)

3.2. K-means dimension re-clustering

K-means is a widely used clustering algorithm. The central idea is to divide each point into clusters,
which are then represented by the nearest cluster center with a given K value and K first-category
center points.Finally, it specify a point and iteratively update the cluster center point until cluster
center does not change much or reaches the specified number of iterations. By the time YOLOv2,
K-means has been used, which can recognize more types of target and has better performance than
YOLOv1. So it has also continued to the current YOLOv5.
The improved YOLOv5 model has one more output scale than the original one. The original 3*3 a
priori anchor frame is no longer applicable. Therefore, it is necessary to re-cluster the improved
YOLOv5 by K-means to make the accuracy recognition and detection better. After many experiments
and research, according to the Avg IOU, it is concluded that when K=12, the final detection accuracy
is better.
The results of the Avg IOU ratio under different K values are shown in Figure 3.

5
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

Fig. 3 Re-clustering results using K-means

The distribution of the candidate boxes after clustering is illustrated in Table 2 below. The
increased feature map scale of 160*160 can be used for the detection of smaller targets, while the
80*80, 40*40, 20*20 in the original YOLOv5 can be used for detecting small, medium or large targets,
by assigning different sizes of a priori boxes, the ability of the network models to detect helmets of
different sizes can be further enhanced.
Tab. 2 Distribution of prior boxes after clustering
Feature map size Anchor frame size
160*160(Smaller scale) (9,11) (12,14) (16,19)
80*80(small scale) (22,26) (29,34) (38,44)
40*40(Medium scale) (49,56) (64,76) (86,101)
20*20(Large scale) (123,142) (188,206) (306,349)

3.3. Loss function

The YOLOv5 uses the binary cross-entropy loss function to compute the loss of category probability
and target confidence score. This paper chooses between GIOU Loss and CIOU Loss through
experimental results, and finally chooses the CIOU Loss as Location loss function, The loss of the
bounding box is equal to 1-CIOU, where the formula of CIOU Loss is:
ρ2 (b,bgt)
ιCIOU = 1 − IOU + c2
+ αγ (1)
4 ωgt w
γ = π2 (arctan hgt − arctan h )2 (2)
γ
α = (1−IOU)+γ (3)

Among them, γ is a parameter that measures the consistency of the aspect ratio, α is a parameter
ωgt w
used to make trade-offs, hgt is the aspect ratio of the bounding box, and h is the aspect ratio of the
predicted frame. CIOU considers the overlap of the frame on the basis of the IOU, and the center The
scale information of distance and aspect ratio makes the final prediction effect better.

4. Experimental results and comparative analysis

The operating system of the experimental computer in this article is Windows10, the CPU model is
Intel(R) Core(TM)i7-7700k [email protected], the GPU model is GeForce RTX 1070, the video memory
size is 8GB, and the memory size is 16GB. All network models are based on Pytorch 1.9, and use
Cuda 11.0 and Cudnn 8.0.4 to accelerate the GPU.

4.1. Data set production and processing

In deep learning, the quality of the data set will greatly affect the quality of the final experimental
results. We have obtained a data set of more than 18,500 operation images of power workers by means

6
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

of web crawlers, of which the training set is 15,900 images, the verification set is more than 2,000, and
the test set is more than 500. The labels are labeled with the Labelme tool. There are two types of
labels, namely "helmet" and "no helmet". The distribution of data set label operation, training set, test
set and data samples are illustrated in Figure 4.

(a) Annotation diagram (b) Train set

(c) Test set (d) Distribution of data samples

Fig. 4 Partial images of the data set
For the sake of reduce the problem of over-fitting during network training, data enhancement was
carried out before training, and methods such as rotation, contrast change, flipping, cropping, and
scaling were used to enhance the self-made data set.

4.2. Network training

Before network training, first configure the hyperparameters of the model, Such as the original
learning rate is lr0=0.01, the final learning rate is lr0*lrf=0.002, the weight decay coefficient is
weight_decay=0.0005, the learning rate momentum is momentum=0.937, the warmup initial
momentum is 0.8, and its bias learning rate is 0.1 and so on. The SGD is used as the optimizer, and the
training period is set to 150 epochs. After configuring these initial hyperparameters for training, we
will get the initial pre-model, that is, the model effect of the original YOLOv5.
After several fine-tunings, the improved network model of us is added, the environment for training
our network model is configured. The K-means dimension is used to re-cluster, and the rectangle is
used to fill the training to accelerate the model inference process. We use the training method of
adding some weights to the images that were not well trained in the previous round, and adjusting the
cosine annealing function value in the hyperparameters and changing the image clipping ratio, flip
direction, rotation angle, and zoom size, learning rate momentum value, mixup coefficient, etc. After
many experiments, we finally got the improved network model.

4.3. Comparison of YOLOv5 initial model and improved model

The performance of the model requires a good evaluation method. For the sake of reduce the problem
of over-fitting during network training, data enhancement was carried out before training, and methods

7
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

such as rotation, contrast change, flipping, cropping, and scaling were used to enhance the self-made
data set. prevent the uneven distribution of the sample targets, we use the precision and the recall rate
to measure. The precision is mainly for the level of prediction results, and the recall rate is mainly for
its own samples. The relevant formula is as follows:
TP
Precision =
TP+FP
(4)
TP
Recall = TP+FN (5)
In the formulas, TP (True positives) can be understood as judging the positive class as the positive
class, that is, the amount that the prediction in the model is correct, and FP (False positive) can be
understood as the negative class being judged as the positive class, that is, the amount that the
prediction in the model is wrong, FN (False Negative) can be understood as a positive class judged as
a negative class, that is, the amount of the model that was originally a positive sample but was missed.
All the experimental results in this article are warmup training. This is to maintain the stability of
the model structure and will not cause oscillation effects due to the high initial learning rate of the
model. After passing the warmup stage, The cosine annealing algorithm is accustomed to update the
learning rate, so as to achieve a better network model. The curve of the relative change of the
YOLOv5 original model and improved model is Precision (accuracy), mAp_0.5, mAp_0.5:0.95, as
shown in Figure 5, 6, and 7, where the higher curve is the consequence of the improved model, and the
lower one is the consequence of original model.

(The x-axis is the period and the y-axis is the precision)

Fig. 5 Precision of YOLOv5 initial model and improved model

8
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

(The x-axis is the period and the y-axis is the mAP_0.5)

Fig. 6 mAp_0.5 of YOLOv5 initial model and improved model

(The x-axis is the period and the y-axis is the mAp_0.5:0.95)

Fig. 7 mAp_0.5:0.95 of YOLOv5 initial model and improved model

It’s observed that that the precision and mAp value of the improved YOLOv5 network model are
higher than those of the original, which can effectively reduce the misdetection rate and missed
detection rate, The comparison of the helmet detection effect of the improved model and the original
model is illustrated in Table 3.
Tab. 3 Comparison of detection performance of YOLOv5 initial model and improved model
Model
Algorithm Parameter/piece Precision/% Recall/% Speed/fps
size/Mb
the initial
YOLOv5 7066239 13.7 92.2 94 32.6

the improved
YOLOv5 7851796 15.3 94.6 97 29.2

9
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

From the table, we can read that the parameters of the improved model are more than those of the
initial one, but the model size is not much different. The improved detection speed is 28.6fps, which is
only 3.4fps less than the original, and it will not have a great impact on the detection. However, the
improved detection precision rate has increased by 2.4% compared to the original, reaching 94.6%,
and the recall rate has increased by 3%, reaching 97%. By the comparison and analysis of the results,
the improved YOLOv5 fully meets the real-time detection requirements, and the detection precision
and recall rate are better than those of the original. Therefore, the improved YOLOv5 network model
in this article is effective in the detection task of whether power workers wear helmets or not.

4.4. Algorithm detection capability comparison

For the sake of fully evaluate and compare the improved YOLOv5 recognition and detection algorithm
in this article, we conducted the following types of experiments:
 The mAp values of the improved YOLOv5 and the original YOLOv5 by using SGD and
Adam are compared, as illustrated in Table 4.The consequence indicate that the SGD
optimizer is superior to the Adam.
 Although the Adam optimizer converges faster, it will make the learning rate converge
quickly and reduce the final mAp value.Therefore, the SGD optimizer is used by default in the
subsequent experiments.
Tab. 4 Comparison of results under different optimizers before and after the improvement
SGD Adam
Algorithm
[email protected] [email protected]:0.95 [email protected] [email protected]:0.95
YOLOv5 0.921 0.557 0.905 0.497
Improved YOLOv5 0.95 0.568 0.921 0.534
 The values of average precision and detection speed between YOLOv3, YOLOv5, and
improved YOLOv5 were compared, and the consequence are illustrated in Table 5.The results
show that the average precision rate of the improved YOLOv5 is 2.2% higher than YOLOv3
and 2.9% higher than YOLOv5, which significantly improves the detection precision.
Although the speed is 3.4fps slower than YOLOv5, it is 10.5fps faster than YOLOv3. The
comprehensive comparison shows that the improved network model has been able to meet
real-time safety helmet detection, and has reached a high precision rate.
Tab. 5 Experimental comparison between different algorithms
Algorithm Average precision/% Detection speed/fps
YOLOv3 92.8 18.7
YOLOv5 92.1 32.6
Improved YOLOv5 95.0 29.2
 The differences between the improved YOLOv5 and the original one for detection under
special conditions are compared. Some examples are shown in Figure 8-13. From left to right,
the input image, the original YOLOv5 detection result and the improved YOLOv5 detection
result are shown. Judging from the comparison of image detection effects in six different
situations, the improved YOLOv5 model has obvious advantages in reducing the missed
detection rate and misdetection rate of object detection in various situations, which improves
the precision of detection. It is advanced in safety helmet detection algorithm of power
workers.

10
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

Fig. 8 The effect of the initial YOLOv5 and the improved model in low light background

Fig. 9 The effect of the initial YOLOv5 and the improved model in the presence of obstruction

Fig. 10 The effect of the initial YOLOv5 and the improved model in the case of missed detection

Fig. 11 The effect of the initial YOLOv5 and the improved model in the case of false detection

Fig. 12 The effect of the initial YOLOv5 and the improved model with remote small targets

11
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

Fig. 13 The effect of the initial YOLOv5 and the improved model with dense small targets

5. Conclusion
Aiming at the poor detection effect and low accuracy of power workers helmets, this paper put
forward an improved YOLOv5 algorithm, which optimizes the feature fusion layer and multi-scale
detection layer, and adds a fusion scale for small target recognition Layer. greatly improving the
detection ability of small targets or even dense small targets. By increasing the size of the feature map
of 160*160, the detection accuracy of the network model for small targets is improved obviously, and
the rate of misdetection and missed detection is reduced. The K-means clustering method is also
accustomed to find the suitable detect candidate anchor frames for small targets such as helmets.
Through comparative experimental analysis, the improved YOLOv5 network model still meets
real-time detection, and the final precision rate is higher, the recall rate is higher, and the stability of
the algorithm is improved obviously.

Acknowledgments
This work was supported by the National Natural Science Foundation of China under Grant
61562025,and Grant 61962019.

References
[1] Zou Z X, Shi Z W, Guo Y H and Ye J P 2019 Object detection in 20 years: A survey J.
Computer Vision and Pattern Recognition
[2] Viola P and Jones M 2001 Robust real-time face detection Proceedings Eighth IEEE
International Conference on Computer Vision (ICCV) pp 747-747
[3] Llorca D F, Arroyo R and Sotelo M A 2013 Vehicle logo recognition in traffic images using
HOG features and SVM Proceedings of International IEEE Conference on Intelligent
Transportation Systems (ITSC) pp 2229-2234
[4] Felzenszwalb P F, Girshick R B, McAllester D and Ramanan D 2010 Object Detection with
Discriminatively Trained Part-Based Models J. IEEE Transactions on Pattern Analysis and
Machine Intelligence 32(9) pp 1627-1645
[5] Liu X H and Ye X N 2014 The application of skin color detection and Hu moment in helmet
recognition J. East China University of Science and Technology (Natural Science Edition)
40(03) pp 365-370
[6] Kido S, Hirano Y and Hashimoto N 2018 Detection and classification of lung abnormalities by
use of convolutional neural network(CNN) and regions with CNN features (R-CNN)
International Workshop on Advanced Image Technology (IWAIT) pp 1-4
[7] Girshick R 2015 Fast R-CNN IEEE International Conference on Computer Vision (ICCV) pp
1440-1448
[8] Ren S, He K, Girshick R and Sun J 2017 Faster R-CNN: Towards Real-Time Object Detection
with Region Proposal Networks IEEE Transactions on Pattern Analysis and Machine
Intelligence vol 39 pp 1137-1149
[9] He L, Zhang X, Ren S and Sun J 2015 Spatial Pyramid Pooling in Deep Convolutional
Networks for Visual Recognition IEEE Transactions on Pattern Analysis and Machine
Intelligence vol 37 pp 1904-1916
[10] Redmon J, Divvala S, Girshick R and Farhadi A 2016 You Only Look Once: Unified,
Real-Time Object Detection IEEE Conference on Computer Vision and Pattern Recognition

12
ICCBDAI-2021 IOP Publishing
Journal of Physics: Conference Series 2171 (2022) 012006 doi:10.1088/1742-6596/2171/1/012006

(CVPR) pp 779-788
[11] Poirson P, Ammirato P, Fu C, Liu W, Kos̆ecká J and Berg A C 2016 Fast Single Shot Detection
and Pose Estimation Fourth International Conference on 3D Vision (3DV) pp 676-684
[12] Redmon J and Farhadi A 2017 YOLO9000: Better, Faster, Stronger IEEE Conference on
Computer Vision and Pattern Recognition (CVPR) pp 6517-6525
[13] Redmon J and Farhadi A 2021 YOLO v3: an incremental improvement
[14] Bochkovshiy A, Wang C and Liao H 2020 YOLOv4: optimal speed and accuracy of object
detection J. Computer Vision and Pattern Recognition
[15] Jocher G 2020 Yolov5 https://github.com/ ultralyc-s/yolov5
[16] Tan M X and Le Q V 2019 EfficientNet: Rethinking Model Scaling for Convolutional Neural
Networks International Conference on Machine Learning vol 97 pp 6105-6114
[17] Huang Z, Zhong Z, Sun L and Huo Q 2019 Mask R-CNN With Pyramid Attention Network for
Scene Text Detection IEEE Winter Conference on Applications of Computer Vision (WACV)
pp 764-772
[18] Wang C, Liao H M, Wu Y, Chen P, Hsieh J and Yeh I 2020 CSPNet: A New Backbone that can
Enhance Learning Capability of CNN IEEE/CVF Conference on Computer Vision and
Pattern Recognition Workshops (CVPRW) pp 1571-1580
[19] Yang J, Fu X, Hu Y, Huang Y, Ding X and Paisley J 2017 PanNet: A Deep Network Architecture
for Pan-Sharpening IEEE International Conference on Computer Vision (ICCV) pp
1753-1761

13