Automatic Score in Archery Target Using Simple Image Processing

Method

Abidul Qohar , Reesa Akbar

and Akhmad Hendriawan

Politeknik Elektronika Negeri Surabaya, Jalan Raya ITS, Surabaya, Indonesia

Keywords: Archery, Automatic Score Detection, Image Processing, OpenCV, Judging System, Camera, System

Development.

Abstract: In this paper we present our developing tool in archery for judging system, which is workable for outdoor and

indoor archery competition. The system used in this tool uses a camera to capture archery targets then the

frames obtained from the camera will be processed using the opencv library to get the score. We do a test

simulation program using opencv and python with video input as an experimental object. We have

successfully collected data from 4 subjects with different brightness settings. The results obtained using this

system for score detection accuracy is 96.55%. This paper introduces a new judging system with high

accuracy and displays the results.

1 INTRODUCTION

Archery is a sport with a long history that train focus

and composure. According to worldarchery.sport, the

first archery competition took place in Finsbury,

England in 1583 which was attended by around 3.000

participants. Furthermore, this sport was officially

included in the Olympics in 1900 for men and 1904

for women. In archery competition, the distance

between the archer and the target board is 70 meters

with the target board tilted 75 degrees to the ground

and the center height of the target board is 1.3 meters

from the ground. On the target there are 10 circles is

the same, from the center yellow, red, blue, black and

white. The determination of the score in the current

archery professional competition uses the Falco Eye

System, which is a laser scanning system to determine

the point of hitting the arrow on the target. This

system is an electronic target using lasers mounted on

the sides. This system is quite accurate, but

unfortunately quite expensive and not easy to move

because it weighs up to 86 kilograms. In amateur and

junior level archery competition, judges and trainers

manually determine the points earned by archers by

either using binoculars or by approaching the target

board to observe and validate the position of the

arrow (Humaid, et al., 2021).

https://orcid.org/0000-0001-7630-7760

In this paper, we would like to implement an

archery score detection system with low memory

usage and simpler method as in, in order to obtain the

most suitable to mount the installation and capture the

whole archer base target. Recently, the computer

vision is a new technology, which can provide more

convenient applications for users. Imaging

recognition is the key technology in vision for various

applications (Hsia, Wang, Cheng, & Chang, 2021).

Many researchers are interested in developing this

system because of its simplicity in determining

archery scores. Most of the papers and journals

published on the internet still detect scores using

images or videos that are processed to get the score

data. In journal published by Thi thi zin, Ikuo oka,

Takuya Sasayama and Shingo ata, the arrow detection

using image approach and the score calculation not

real time. The low computational complexity and the

easiness of implementation are the key advantages of

proposed method in the journal (Zin, et al., 2013).

Another journal discussing the topic of score

detection is published by Raymond Parag. The

journal discusses score detection in archery with a

video approach. The frame processed continuously to

predict the position of the arrow on the archery target

board based its color (Parag, 2017).Various methods

and techniques using OpenCV are discussed and

applied to detect and score the arrows. The target is

488

Qohar, A., Akbar, R. and Hendriawan, A.

Automatic Score in Archery Target Using Simple Image Processing Method.

DOI: 10.5220/0011816400003575

In Proceedings of the 5th International Conference on Applied Science and Technology on Engineering Science (iCAST-ES 2022), pages 488-492

ISBN: 978-989-758-619-4; ISSN: 2975-8246

detected by a color-based approach. The perspective

of the target is corrected by a homographic

transformation matrix and arrows are detected and

scored using the Hough lines Transform in

combination with a point in contour test.

2 IMAGE PROCESSING

TECHNIQUE

This paper provide research about score detection in

archery using image processing technique that

programmed by python language. The method used

in this paper divided in many process, there are

Background Subtraction, Morphological Operation,

Arrow Detection, Target Detection, and Scoring

Calculation. Detailed information about how we

calculate the score in archery using image processing

explained in Figure 1.

Figure 1: Method Proposed for Automatic Scoring System.

2.1 Background Subtraction

Background subtraction is an approach used to detect

moving objects in a sequence of frames from static

cameras. The base in this approach is that of detecting

moving objects from the difference between the

current frame and reference frame, which is often

called ‘Background Image’ or ‘Background Model’.

This task is the fundamental step in many visual

surveillance applications for which background

subtraction is a suitable solution which provide a

good compromise in terms of quality of detection and

computation time (Bouwmans & Garcia, 2019).

Frame difference is used to estimate the background

image by comparing the previous frame with the

current frame. This approach can be used when

segment motion-based objects such as arrow

movement, cars, pedestrians etc. And it is very

sensitive to threshold value settings. Therefore there

are limitation usage depending on object structure,

frame rate and global threshold. Figure 2 is some

cases of this approach based on threshold values.

Threshold = 30

Threshold = 50

Threshold = 80

Threshold = 100

Figure 2: Approach based on threshold value.

From Figure 2 when threshold value is set to 30, the

circle is not visible compared to when threshold value

at 100, the circle look clearer and easier to detect.

The common method used for background

subtraction is Mixture of Gaussians. The Mixture of

Gaussians is a mixture of k Gaussians distribution

models for each background pixel with values for k

within 3 and 5. The inventor assumes that different

distributions each represent the different background

colors and foreground colors. The weight of each one

of those used distributions on the model is

proportional to the amount of time each color stays on

pixel. Therefore when the weight of pixel distribution

is low, the pixel is classified as a foreground pixel.

2.2 Morphological Operations

Morphology is a vast extent of image processing

operations that modifies the images based on shapes.

It is considered to be one of the data processing

methods useful in image processing (Priya & Kadhar,

2017). Morphological operations apply a structuring

element to an input image, creating an output image

of the same size. In a morphological operation, the

value of each pixel in the output image is based on a

comparison of the corresponding pixel in the input

image with its neighbors. The most basic

morphological operations are dilation and erosion.

Dilation adds pixels to the boundaries of objects in an

image, while erosion removes pixels on object

boundaries. The number of pixels added or removed

from the objects in an image depends on the size and

shape of the structuring element used to process the

Automatic Score in Archery Target Using Simple Image Processing Method

489

image. In the morphological dilation and erosion

operations, the state of any given pixel in the output

image is determined by applying a rule to the

corresponding pixel and its neighbors in the input

image. The rule used to process the pixels defines the

operation as a dilation or an erosion. Table 1 is lists

of the rules for both dilation and erosion (Priya &

Kadhar, 2017).

Table 1: Rules of dilation an erosion.

Operation Rules

Dilation The value of the output pixel is

the maximum value of all pixels in the

neighborhood. In a binary image, a

pixel is set to ‘1’ if any of the

neighboring pixels have the value ‘1’.

Morphological dilation makes objects

more visible and fills in small holes in

objects. The lines appear thicker, and

filled shapes appear lar

er.

Erosion The value of the output pixel is

the minimum value of all pixels in the

neighborhood. In a binary image, a

pixel is set to ‘0’ if any of the

neighboring pixels have the value ‘0’.

Morphological erosion remove

floating pixels and thin lines so that

only substantive objects remain. The

remaining lines appear thinner and

shapes appear smaller.

2.3 Arrow Detection

Arrow detection in this paper using data obtained

from background subtraction and hough line method.

When another object enters the frame, that object will

be considered as an arrow that sticks to the target of

the archery. From the detection of arrows using the

hough line method, two coordinates will be obtained

which are the coordinates of the start of the line

(assume as “a” coordinate) and the end of the line

(assume as “b” coordinate). After we get the data, the

next step is to look for the contours of the arrows as

parameters to capture images or only 1 frame is

processed, this is to minimize the use of cpu so that it

is not too big. Early approaches to contour detection

aim at quantifying the presence of a boundary at a

given image location through local measurements

(Arbelaez, Maire, Fowlkes, & Malik, 2011).

To determine the end of the arrowhead, the

approximation of the coordinates of center circle is

used. If coordinate "a" is closer to the center of the

target than coordinate "b", then point "a" is the

arrowhead and point b is the arrow's tail, and vice

versa. The center of the archery target is obtained

based on the target detection method described in

section 2.4. This arrow detection test is very crucial

in determining the score in this scoring process. The

accuracy of the arrowhead tip detection will affect the

accuracy of the score detection obtained.

Figure 3: Arrow head detection using approximation of the

coordinates of center circle.

2.4 Target Detection and Scoring

Calculation

After the arrow head point is found or detected, the

system need to find in which ring this point is located

(Danielescu, 2021). Archery target testing aims to

detect score circles which will be used as an important

component in detecting archery scores. The method

used in detecting circles on archery targets is the

hough circle method. From this method, the radius

and coordinates of the center point of the circle will

be obtained. To get more accurate results, it is

necessary to first set the threshold of the frame. The

score circles obtained from detection using the hough

circle method will later be compared with the position

of the tip of the arrow obtained previously and then

the score for the arrow will be obtained.

Calculating the archery score is obtained by

comparing the results of the previous steps. That is by

measuring the radius from the arrowhead to the

coordinates of the center of the circle compared to the

radius of each circle.

3 EXPERIMENTS AND

SIMULATIONS RESULT

In this experiment and simulation, video input is used

to detect scores on the archery target board. There are

4 arrows that are the object of the experiment in

determining the accuracy from the method proposed

in this paper. Then experiment by giving different

light conditions to see the difference in accuracy

obtained.

iCAST-ES 2022 - International Conference on Applied Science and Technology on Engineering Science

490

3.1 Score Detection

Score detection is the final process in this paper. The

calculation used in determining the score in this paper

uses data obtained from the previous process. The

score calculation is carried out by comparing

coordinates of the arrowheads in the radius of rings.

The results are shown in Figure 4 until Figure 7.

Figure 4: Score detection on the first arrow by calculate the

radius from the arrowhead to the center of circle.

Figure 5: Score detection on the second arrow by calculate

the radius from the arrowhead to the center of circle.

Figure 6: Score detection on the third arrow by calculate the

radius from the arrowhead to the center of circle.

Figure 7: Score detection on the fourth arrow by calculate

the radius from the arrowhead to the center of circle.

Table 2: Score detection accuracy.

Arrow

umbe

Score Actual Score Detecte

1 7 7

2 9 9

3 7 6

4 7 6

Total 30 28

Accurac

93.33%

3.2 Various Brightness

This experiment aims to see whether the intensity of

the light given affects the accuracy of detecting scores

on archery targets. There are 2 experimental objects

with different light intensity, among others, by

increasing the brightness of +10, and +40. From this

experiment, the following results were obtained.

Figure 8: Brightness increasing by +10.

Table 3: Score detection accuracy with brightness

increasing by +10.

Arrow Numbe

Score Actual Score Detecte

1 7 7

2 9 9

3 7 6

4 7 7

Total 30 28

Accurac

96.66%

Automatic Score in Archery Target Using Simple Image Processing Method

491

Figure 9: Brightness increasing by +40.

Table 4: Score detection accuracy with brightness

increasing by +40.

Arrow Numbe

Score Actual Score Detecte

1 7 7

2 9 8

3 7 6

4 7 4

Total 30 25

Accurac

83.33%

4 CONCLUSIONS

In this paper, we have proposed score detection in

archery target using simple image processing method.

The experimental results show that in terms of

performance, the program for score detection is

running well. However, the light conditions need to

be adjusted to get higher accuracy. This research will

be very well implemented in stable lighting

conditions. This paper can help referees determine

scores in archery so that the time spent in judging can

be more effective. In future, the automatic lightning

calibration will be included to increase the accuracy

of the system.

REFERENCES

Arbelaez, P., Maire, M., Fowlkes, C., & Malik, J. (2011).

Contour Detection and Hierarchical Image

Segmentation. IEEE Transactions on Pattern Analysis

and Machine Intelligence, (pp. 898-916).

doi:10.1109/TPAMI.2010.161

Bouwmans, T., & Garcia, B. (2019). Background

Substraction in reak Applications: Challenges, Current

Models and Future Directions. Cornell University.

doi:https://doi.org/10.48550/arXiv.1901.03577

Chudasama, D., Patel, T., & Joshi, S. (2015). Image

Segmentation using Morphological Operations.

International Journal of Computer Applications, 117,

pp. 16-19. Bharuch. doi:10.5120/20654-3197

Danielescu, L. A. (2021). OnTarget: An Electronic Archery

Scoring System. Cornell University.

doi:https://doi.org/10.48550/arXiv.2104.01622

Hsia, S.-C., Wang, S.-H., Cheng, W.-C., & Chang, C.-Y.

(2021). Intelligent Blowgun Game Scoring Recognition

System Based on Computer Vision. IEEE Access, 9, pp.

73703-73712.

doi:https://doi.org/10.1109/ACCESS.2021.3081457

Humaid, H., Pelana, R., Wendy, C.-S., Winata, B., Yulianti,

E., Wattimena, F. Y., & Artono, E. (2021). Validity of

the Scoring System Technology for Detecting Points in

Archery. Journal of Physical Education and Sport,

21(3), 1520-1524. doi:10.7752/jpes.2021.03193

Parag, R. (2017). Sequential Recognition and Scoring of

Archery Shots. Leiden University. Retrieved from

https://theses.liacs.nl/pdf/RaymondParag.pdf

Priya, M. S., & Kadhar, N. (2017). Effective Morphological

Image Processing Techniques and Image

Reconstruction. National Conference on "Digital

Transformation-Challenges and Outcomes" (ASAT in

CS'17). Bangalore.

Zin, T. T., Oka, I., Sasayama, T., Ata, S., Watanabe, H., &

Sasano, H. (2013). Image processing Approach to

automatic Scoring System for Archery Targets. In 2013

Ninth International Conference on Intelligent

Information Hiding and Multimedia Signal Processing

(pp. 259-262). IEEE. doi:https://doi.org/10.1109/IIH-

MSP.2013.73

iCAST-ES 2022 - International Conference on Applied Science and Technology on Engineering Science

492