CC BY
7Technobius, 2023, 3(1), 0032, DOI: https://doi.org/10.54355/tbus/3.1.2023.0032
Technobius
https://technobius.kz/
e-ISSN 2789-7338
Article
Methodology for determining coordinate points using automated software and
aircraft
Zhassulan Kuzbakhov, Shyngys Zharassov*
Department of Civil Engineering, L.N. Gumilyov Eurasian National University, Astana, Kazakhstan
Correspondence: zhshzh95@gmail.com
Abstract. Engineers are currently facing questions about the use of geographic information systems or software to implement projects in a short time. The problem with using geographic information systems in construction is the relevance of the available data. An example is open sources with satellite images. This problem appeared even before satellite-positioning systems emerged. In this connection, the purpose of this article is to find the deviation of source points when performing photogrammetry with marker detection in Agisoft PhotoScan software. This method of determining coordinates using a single point and its correlation on the ground is applicable in the case of rapid calculations, where the volumes of earth masses are large enough and do not require increased accuracy at the stage of approximate calculations. As a result of the comparison of traditional and automated methods of definition of coordinates on the ground has been found an essential distinction, both in total values and in time spent for the definition of points of coordinates. The considerable difference revealed by the results of the comparison of coordinates is presented in the table as a color gradation. The average deviation between known coordinates and coordinates obtained in Agisoft PhotoScan by axes was: X=0.87%, Y=0.45%, Z=0.12%.
Keywords: Agisoft PhotoScan, surveying, automation, photogrammetry, geodesy, topographic mapping.
1. Introduction
The use of digital tools in automating construction processes is rapidly advancing, despite the challenges posed by the complexity and diversity of these processes. There is a growing interest in leveraging geographic information systems and software to streamline project implementation, which raises questions for engineers about how best to apply these tools within tight timeframes. However, integrating such tools can lead to errors and deviations caused by human mistakes or insufficient information on their practical application [1].
When it comes to using geographic information systems in construction, the main challenge is the quality and relevance of the available data. For instance, open sources like satellite images may be of limited use due to their quality, coverage, and the current state of urbanization in the area being assessed. This can make it difficult to accurately evaluate the situation for construction projects [2].
Although surveying a plane with a GPS receiver is relatively straightforward, those who use optical and satellite instruments should be aware of the potential challenges associated with these tools. Attempting to integrate traditional and satellite surveying instruments can lead to several points for consideration [3], including:
- The coordinate system used when taking pictures;
- Relativity surface;
- Scale factor of the projection;
- Correction for projection height, etc.
This issue has been a challenge even prior to the advent of satellite positioning systems. For example, when using high-precision total stations of the Leica series, it was found that the deviation was about 0.2 meters, factoring in the coordinate system adopted in the region being surveyed. This deviation occurs despite considering the installation of the device and the position of the reflector. It's important to note that Leica GPS receivers typically determine the coordinates of points in the WGS-84 geodetic coordinate system. However, in practice, the UTM32 coordinate system is used when converting to plane coordinates, which establishes the relationship between the ellipsoid surface part and the plane coordinates in the projection. Different results can be obtained by using a Leica TPS total station (Figure 1), which determines the coordinates over a peg. Measurements taken with a reflector can also have significant errors if the ellipsoid surface is not taken into account [4].
/ '~~ > j / \
^ " " \ \ / / / / / / Of / / _L_L_ N 1 \ N
o ""So--"' ioo__ \ i* a— * \ Jo / ^Ni00 / / 250 \l 300 350 / -" /
♦-K. \ s /
—Min-Confidence interval of deviation
—*—Max-Confidence interval of deviation
---Average deviation
Horizontal circle setting, degrees
Figure 1 - Tachymeter accuracy Leica TS006 [4]
Moreover, there are software programs available that can automate surface breakdowns by correlating points based on surface topography. However, deviations in these measurements can lead to unpredictable consequences, as minimum and maximum values can significantly affect the accuracy of the results [5]. The accuracy of the measuring instrument used, as well as other factors such as image quality, weather conditions, and the skill of the operator flying drones or other aircraft, are critical considerations in automating the surveying process [6].
Therefore, the objective of this article is to determine the deviation of baseline points when conducting photogrammetry using marker detection in the Agisoft PhotoScan software. Agisoft PhotoScan is primarily an autonomous software tool that can conduct photogrammetric processing of digital images and generate three-dimensional spatial data for use in Geographic Information Systems (GIS). Therefore, the authors of this article showcase an example of executing topographic surveys using this software. The example pertains to a ground excavation site, which was utilized for embankment or excavation during the construction of residential complexes [7].
2. Methods
The method used in this study to determine coordinates involves using a single point and its correlation on the ground, which is suitable for quick calculations when the volume of earth masses is sufficiently large and high accuracy is not required at the stage of approximate calculations.
For this study, Agisoft PhotoScan software was utilized. In addition to the advantages previously mentioned, an additional benefit of using stand-alone software is its flexibility and adaptability to various systems for calculations.
The drone used in this study for photogrammetry is the Quadcopter DJI Mini 3 Pro. The selection of this model was based on the manufacturer's specifications, with special emphasis on the matrix size and resolution of the captured frames. Additionally, the number of frames captured and the time of day of shooting are critical factors to consider. A higher number of frames captured
enables the accurate construction of a 3D model, while the absence of shadow zones facilitates the calculation of surface relief [8].
The survey process is illustrated in Figure 2 below. The drone's trajectory for mapping the terrain can follow a continuous circular motion around the center or the edge of the surveyed area, or move along a path from the beginning to the end of the polygon within the defined boundaries of the area. The captured images are aligned with a certain degree of accuracy, where the orientation of each image is tied to the angle of view of the drone's camera, resulting in the creation of a point cloud.
2cf1 " " Bp • • 9 % • A. 4 /'> « ■■ ■ * ♦ * • * ■•?!»* +
=^=u—7 j§ir j§r шг jgf jr jgr jar
Figure 2 - The process of building a point cloud for polygon photogrammetry in Agisoft PhotoScan
The process of creating a model for marker detection involves processing the point cloud, which is a time-consuming task. However, the output is a model of high quality that can be used to detect markers. These markers, which are placed on the boundary of the area being defined, contain the initial coordinates of reference points. In the field, these reference points are determined at the survey site by marking. The markers can take the form of a "+" mark, such as a plus sign, which is easily visible in an aerial survey, or an object with a clear center, as shown in Figure 3.
И 1 po* — Apaeft ftiotoScin РтеЛеякмм! - " U
¡.I - •• ..•♦•»V • • . * а и m u. +
4
Figure 3 - Identification of markers on the territory of the polygon in the Agisoft PhotoScan
environment
3. Results and Discussion
After comparing traditional and automated methods of determining ground coordinates, it was discovered that there were notable discrepancies in both the final values and the time taken to determine coordinate points. Previous research exploring comparative deviation analysis, focused on creating point clouds via laser and digital photogrammetry, is discussed in [9]. The accuracy of point positioning by computer programs is analyzed in [10], where the coordinates of points obtained through tachymetric survey are used as the reference system by the authors.
Figure 4 below shows the model subjected to linear transformations.
ssr mmmm mm
® * * <w«(ji>wit^|.»«c MMMBj'Oitftt'l.wi.Mij»« MMjiiiMiwaiM uMWj*mnjn.«UK uwm.,i)Wiin'i.aiK M4<MjX<MOi' i.«i.irK
Figure 4 - Building a vertical leveling profile in Agisoft PhotoScan according to markers
The similarity transformation model is derived from 7 parameters: 3 translation parameters, 3 rotation parameters, 1 stretching and compression parameter). This approach solves only linear distortions while nonlinear distortions which are also present in the model can be the reason for further errors in the model georeferencing and calculations. To reduce influence of nonlinear distortions markers or reference points with known coordinates were used.
Table 1 below shows the deviation between traditional point coordinates (X1, Y1, Z1) and automated (X2, Y2, Z2) methods in Agisoft PhotoScan.
Table 1 - Deviation of coordinates
№ X1 X2 Xdiff Incl% Y1 Y2 Ydiff Incl% Z1 Z2 Zdiff Incl%
1 47.8771 48.8834 1.0063 2.10% 67.5170 68.5548 1.0378 1.54% 356.9000 357.9265 1.0265 0.29%
2 47.8771 48.9085 1.0314 2.15% 67.5170 67.5309 0.0139 0.02% 357.2000 357.2502 0.0502 0.01%
3 47.8771 48.8904 1.0133 2.12% 67.5170 67.5378 0.0209 0.03% 356.5000 356.5235 0.0235 0.01%
4 47.8771 47.9010 0.0240 0.05% 67.5170 67.5509 0.0339 0.05% 356.7000 356.7276 0.0276 0.01%
5 47.8771 47.9171 0.0400 0.08% 67.5170 67.5235 0.0065 0.01% 356.2000 357,2477 1,0477 0,29%
6 47,8771 47,8840 0,0069 0,01% 67,5170 67,5432 0,0262 0.04% 357.0000 357.0320 0.0320 0.01%
7 47.8771 47.9259 0.0488 0.10% 67.5170 68.5227 1.0057 1.49% 356.3000 356.3364 0.0364 0.01%
8 47.8771 47.9240 0.0470 0.10% 67.5170 67.5344 0.0174 0.03% 356.8000 357.8267 1.0267 0.29%
9 47.8771 48.9218 1.0447 2.18% 67.5170 67.5628 0.0458 0.07% 356.6000 357.6500 1.0500 0.29%
10 47.8771 48.8881 1.0110 2.11% 67.5170 67.5474 0.0303 0.04% 357.2000 357.2447 0.0447 0.01%
11 47.8771
12 47.8771
13 47.8771
14 47.8771
15 47.8817
16 47.8818
17 47.8815
18 47.8812
19 47.8810
20 47.8809
21 47.8809
22 47.8810
23 47.8810
24 47.8809
25 47.8809
26 47.8809
27 47.8808
28 47.8805
29 47.8801
30 47.8797
31 47.8793
32 47.8792
33 47.8795
34 47.8798
35 47.8800
36 47.8802
37 47.8801
38 47.8800
39 47.8800
40 47.8800
41 47.8801
42 47.8801
43 47.8801
44 47.8801
45 47.8801
46 47.8801
47 47.8800
48 47.8797
49 47.8792
50 47.8788
51 47.8787
52 47.8789
53 47.8791
54 47.8793
55 47.8794
56 47.8793
57 47.8793
58 47.8793
47.8801 48.8946 47.9170 48.9023 48.9001 47.8932 47.9037 47.9295 47.9173 47.8956 47.8950 47.8855 47.8849 48.9207 47.9215 48.8988 48.9199 47.8847 47.9012 48.9117 47.9150 48.8821 47.8830
48.8924 48.9268 47.8893 47.9178
47.8879 47.9297
47.8880 47.9212 47.9189 47.8880 48.9036
47.8925
47.8932
47.8933 48.8975 47.8908 47.9000 48.9074 47.9252 48.9035 48.8923 48.9292 47.9194 47.9292 47.8832
0.0031 0.01%
1.0175 0.0399 1.0252 1.0185 0.0115 0.0222 0.0483 0.0363
2.13% 0.08% 2.14% 2.13% 0.02% 0.05% 0.10% 0.08%
0.0147 0.0140 0.0045 0.0039
0.03% 0.03% 0.01% 0.01%
1.0397 0.0406 1.0179 1.0391 0.0042 0.0211 1.0320 0.0356 1.0028
2.17% 0.08% 2.13% 2.17% 0.01% 0.04% 2.16% 0.07% 2.09%
0.0036 0.01%
1.0126 1.0468 0.0091
2.11% 2.19% 0.02%
0.0377 0.08%
0.0079 0.02%
0.0497 0.10%
0.0079 0.02%
0.0412 0.0389 0.0079 1.0235 0.0124 0.0131 0.0134 1.0178 0.0115 0.0212 1.0287 0.0464 1.0244 1.0130 1.0498 0.0401 0.0499
0.09% 0.08% 0.02% 2.14% 0.03% 0.03% 0.03% 2.13% 0.02% 0.04% 2.15% 0.10% 2.14% 2.12% 2.19% 0.08% 0.10%
0.0039 0.01%
67.5170 67.5170 67.5170 67.5170 67.5212 67.5205 67.5198 67.5192
67.5187 67.5181 67.5175
67.5168
67.5162
67.5156
67.5150
67.5144 67.5138 67.5132 67.5130 67.5130 67.5134 67.5140
67.5145 67.5148
67.5151
67.5157
67.5163
67.5169 67.5175 67.5181
67.5188
67.5194 67.5200
67.5207 67.5214 67.5220 67.5227
67.5231
67.5232 67.5229 67.5223 67.5217 67.5212
67.5208 67.5202
67.5195 67.5188 67.5181
67.5482 67.5557 67.5515 67.5509 67.5470 67.5551
0.0312 0.0387 0.0345 0.0339 0.0258 0.0346
68.5253 68.5481 68.5557
67.5601 67.5570 67.5617
67.5662 67.5656 67.5491 67.5177 68.5253
67.5457 67.5453 67.5615 67.5262 67.5199 67.5628
68.5507 68.5276 68.5386
67.5306 67.5415 67.5552 67.5326
.5565
67.5338 67.5462
68.5583 68.5394
67.5401
68.5328
67.5451 67.5335 67.5487 67.5382 67.5285 67.5703
67.5540 68.5230 67.5221
0.05% 0.06% 0.05% 0.05% 0.04% 0.05%
1.0055 1.0288 1.0370
1.49% 1.52% 1.54%
0.0420 0.0395 0.0448
0.06% 0.06% 0.07%
68.5228 1.0066 1.49%
0.0506 0.0507 0.0348
0.07% 0.08% 0.05%
0.0039 1.0121
0.01% 1.50%
0.0327 0.0323 0.0481
0.05% 0.05% 0.07%
0.0122 0.0055
0.02% 0.01%
0.0480 0.07%
1.0356 1.0120 1.0223 0.0136 0.0239 0.0371
1.53% 1.50% 1.51% 0.02% 0.04% 0.05%
0.0138 1.0371 0.0138 0.0255
0.02% 1.54% 0.02% 0.04%
1.0369 1.0174 0.0174 1.0097
1.54% 1.51% 0.03% 1.50%
0.0219 0.0106
0.03% 0.02%
0.0264 0.0165 0.0072
0.04% 0.02% 0.01%
0.0495 0.07%
5.5319 1.0117 1.50%
0.0345 1.0041 0.0040
0.05% 1.49% 0.01%
356.9000 356.9000 356.9000 357.2000 517.7000 513.8000 514.1000 519.1000 523.0000 519.7000 514.3000 515.5000 516.3000 514.6000 514.9000 516.2000 518.9000 518.1000 514.1000 511.4000 510.1000 509.0000 507.8000 511.4000 517.0000 519.6000 514.1000 517.6000 518.2000 519.5000 523.0000 525.1000 522.3000 519.1000 523.5000 525.1000 526.8000 526.1000 522.4000 518.5000 515.4000 513.9000 518.8000 527.0000 523.8000 521.2000 524.0000 525.2000
356.9422 0.0422 0.01%
356.9248 0.0248 0.01%
356.9343 0.0343 0.01%
358.2510 1.0510 0.29%
519.7231 2.0231 0.39%
513.8261 0.0261 0.01%
514.1107 0.0107 0.00%
519.1238 0.0238 0.00%
524.0058 1.0058 0.19%
520.7050 1.0050 0.19%
514.3113 0.0113 0.00%
515.5459 0.0459 0.01%
517.3044 1.0044 0.19%
515.6264 1.0264 0.20%
514.9234 0.0234 0.00%
517.2142 1.0142 0.20%
518.9173 0.0173 0.00%
520.1299 2.0299 0.39%
515.1421 1.0421 0.20%
511.4121 0.0121 0.00%
511.1148 1.0148 0.20%
510.0468 1.0468 0.21%
508.8086 1.0086 0.20%
511.4267 0.0267 0.01%
517.0330 0.0330 0.01%
519.6370 0.0370 0.01%
516.1508 2.0508 0.40%
518.6193 1.0193 0.20%
518.2295 0.0295 0.01%
520.5162 1.0162 0.20%
524.0091 1.0091 0.19%
526.1468 1.0468 0.20%
522.3347 0.0347 0.01%
519.1506 0.0506 0.01%
523.5495 0.0495 0.01%
526.1498 1.0498 0.20%
527.8038 1.0038 0.19%
526.1165 0.0165 0.00%
522.4164 0.0164 0.00%
518.5383 0.0383 0.01%
516.4364 1.0364 0.20%
514.9192 1.0192 0.20%
519.8361 1.0361 0.20%
528.0124 1.0124 0.19%
523.8103 0.0103 0.00%
522.2282 524.0398 528.2248
1.0282 0.0398
0.20% 0.01%
3.0248 0.58%
59 47.8793
60 47.8793
б1 47.8793
б2 47.8793
63 47.8793
б4 47.8793
б5 47.8792
бб 47.8788
б7 47.8784
б8 47.8779
б9 47.8777
7G 47.877б
71 47.8779
72 47.8782
73 47.8785
74 47.878б
75 47.8785
7б 47.8784
77 47.8784
78 47.8785
79 47.8785
8G 47.8785
81 47.8784
82 47.8785
83 47.8785
84 47.8785
85 47.8784
8б 47.8781
87 47.8777
88 47.8773
89 47.8773
9G 47.8774
91 47.877б
92 47.8777
93 47.8777
94 47.8777
95 47.8777
9б 47.8777
97 47.8777
98 47.8777
99 47.8777
100 47.8777
1G1 47.8777
102 47.877б
103 47.8773
104 47.87б9
1G5 47.87б4
106 47.87б1
47.91б9 47.9G67
47.8961 48.8959
47.8951 47.9154 47.8827 47.89G3 48.887G 49.9242 48.9171 48.8997 47.889б 47.9158
48.8962 47.8837 48.899б 47.9G91
48.8963
47.8952 47.8878 48.9G18
48.8978 47.881б 48.92бб 47.9192 47.914G 47.899G
47.8979 47.9G68 47.9G63 47.9283 47.9G33 48.9G81 47.8878 47.8822 47.8957 47.89G9 47.9227 47.929G 48.8983 48.9G57 47.9GG1 48.8898 47.8843 48.8985 49.8932 48.9G68
G.G376 G.G274 G.G168 1.G166 G.G158 G.G361
G.G8% G.G6% G.G4% 2.12% G.G3% G.G8%
G.GG35 G.G115
1.GG86
2.G463 1.G395 1.G221 G.G117
G.G1% G.G2% 2.11% 4.27% 2.17% 2.13% G.G2%
G.G376 1.G177 G.GG51 1.G211 G.G3G7 1.G179 G.G167 G.GG93 1.G234 1.G194 G.GG31 1.G482 G.G4G7 G.G356 G.G2G9 G.G2G2 G.G295 G.G29G G.G5G8 G.G257 1.G3G4 G.G1G1 G.GG46 G.G18G G.G132 G.G45G G.G512 1.G2G5 1.G28G G.G224 1.G122 G.GG7G
1.G216
2.G168 1.G3G7
G.G8% 2.13% G.G1% 2.13% G.G6% 2.13% G.G3% G.G2% 2.14% 2.13% G.G1% 2.19% G.G9% G.G7% G.G4% G.G4% G.G6% G.G60/o G.11% G.G5% 2.15% G.G2% G.G1% G.G4% G.G3% G.G9% G.11% 2.13% 2.15% G.G5% 2.11% G.G1% 2.13% 4.21% 2.15%
б7.5174
67.5168 б7.51б1
67.5155 б7.5148 б7.5142 б7.5135 67.513G б7.5128 67.513G 67.5134 67.5140
67.5144 67.5147 67.5151
67.5156 б7.51б3
67.5169 67.5174 67.5180 67.5186 67.5193
67.5200
67.5206 б7.5212 б7.5218 б7.5224 б7.5228 б7.5228 б7.5224 б7.5218 б7.5212
67.5207
67.5201 67.5195 67.5189 67.5182 67.5176
67.5170 б7.51б3
67.5157 67.5151
67.5145 б7.5139 67.5133
67.5131
67.5132 67.5137
67.5514 67.5470
0.0340 0.0303
0.05% 0.04%
67.5465 67.5375 67.5614 67.5586 67.5240 67.5250 67.5602
67.5236 0.0076 0.01%
68.5212 1.0058 1.49%
67.5374 0.0226 0.03%
67.5367 0.0225 0.03%
67.5402 0.0267 0.04%
67.5183 0.0052 0.01%
68.5520 1.0392 1.54%
67.5605 0.0476 0.07%
68.5548 1.0414 1.54%
67.5509 0.0369 0.05%
67.5550 0.0406 0.06%
67.5396 0.0249 0.04%
67.5195 0.0044 0.01%
68.5368 1.0212 1.51%
67.5403 0.0240 0.04%
67.5538 0.0369 0.05%
68.5362 1.0188 1.51%
67.5225 0.0045 0.01%
68.5608 1.0422 1.54%
68.5669 1.0476 1.55%
67.5333 0.0133 0.02%
68.5364 1.0158 1.50%
68.5350 1.0138 1.50%
67.5366 0.0148 0.02%
67.5578 0.0354 0.05%
67.5277 0.0049 0.01%
67.5567 0.0339 0.05%
68.5406 1.0182 1.51%
67.5608 0.0390 0.06%
67.5248 0.0036 0.01%
67.5596 0.0389 0.06%
67.5565 0.0364 0.05%
67.5222 0.0027 0.00%
67.5265 0.0076 0.01%
67.5389 0.0206 0.03%
68.5394 1.0217 1.51%
68.5631 1.0461 1.55%
67.5333 0.0169 0.03%
67.5269 0.0111 0.02%
0.0314 0.0230 0.0475 0.0452
0.05% 0.03% 0.07% 0.07%
0.0110 0.0118
0.02% 0.02%
0.0465 0.07%
522.9GGG 521.4000 522.6000 523.8GGG 523.1000 521.9000 521.1000 518.9000 514.7000 5G9.7GGG 5G5.8GGG 5G4.3GGG 5G5.GGGG 511.2000 516.5000 515.8000 513.9000 517.1000 518.0000 517.3000 519.9000 515.5000 516.3000 52G.GGGG 521.4000 520.1000 519.9000 516.9000 5G7.9GGG 5G4.7GGG 5G9.GGGG 5G9.5GGG 513.8000 519.0000 518.5000 52G.3GGG 521.5000 519.9000 517.7000 517.7000 518.8000 517.6000 517.0000 515.3000 511.2000 506.2000 5G4.8GGG 5G4.7GGG
523.9322 521.4080 523.6199 524.8058 523.1313 521.9084 522.1182 518.9364 515.7388 510.7340 506.8484 504.3095 505.0104 513.2402 517.5420 515.8089 514.9069 517.1283 518.0403 517.3493 519.9151 515.5244 517.3090 522.0214 521.4258 520.1106 519.9459 517.9433 508.9377 505.7085 509.0043 510.5238 513.8052 520.0415 518.5161 521.3109 521.5281 520.9314 517.7328 518.7373 519.8412 517.6168 517.0477 515.3031 511.2159 506.2144 504.8493 504.7240
1.0322
G.2G%
0.0080 0.00%
1.0199 1.0058 0.0313
1.0182 0.0364 1.0388 1.0340 1.0484
0.0095 0.0104 2.0402 1.0420
1.0069 0.0283 0.0403 0.0493
0.0151 0.0244 1.0090 2.0214 0.0258 0.0106
0.0459 1.0433 1.0377 1.0085
0.0043
1.0415 0.0161 1.0109 0.0281 1.0314 0.0328 1.0373 1.0412 0.0168
0.0031 0.0159 0.0144
0.20% 0.19% 0.01%
0.0084 0.00%
0.20% 0.01% 0.20% 0.20% 0.21%
0.00% 0.00% 0.40% 0.20%
0.0089 0.00%
0.20% 0.01% 0.01% 0.01%
0.00% 0.00% 0.20% 0.39% 0.00% 0.00%
0.01% 0.20% 0.20% 0.20%
0.00%
1.0238 0.20%
0.0052 0.00%
0.20% 0.00% 0.19% 0.01% 0.20% 0.01% 0.20% 0.20% 0.00%
0.0477 0.01%
0.00% 0.00% 0.00%
0.0493 0.0240
0.01% 0.00%
107 47.8760 48.9202 1.0442 2.18% 67.5143 67.5414 0.0271 0.04% 504.8000 505.8174 1.0174 0.20%
108 47.8762 48.8832 1.0069 2.10% 67.5148 68.5347 1.0200 1.51% 506.0000 507.0223 1.0223 0.20%
109 47.8765 48.8858 1.0092 2.11% 67.5151 68.5238 1.0087 1.49% 509.3000 510.3066 1.0066 0.20%
110 47.8768 47.8932 0.0164 0.03% 67.5155 67.5433 0.0278 0.04% 514.3000 514.3387 0.0387 0.01%
111 47.8769 47.8948 0.0179 0.04% 67.5160 68.5544 1.0384 1.54% 513.6000 513.6049 0.0049 0.00%
112 47.8768 47.8995 0.0227 0.05% 67.5166 67.5477 0.0310 0.05% 512.2000 513.2256 1.0256 0.20%
113 47.8768 47.9158 0.0391 0.08% 67.5172 68.5326 1.0154 1.50% 514.7000 515.7281 1.0281 0.20%
114 47.8768 47.9087 0.0319 0.07% 67.5179 68.5479 1.0300 1.53% 513.7000 513.7311 0.0311 0.01%
115 47.8769 47.8816 0.0048 0.01% 67.5185 67.5247 0.0061 0.01% 513.8000 513.8040 0.0040 0.00%
116 47.8769 48.9042 1.0273 2.15% 67.5192 68.5669 1.0477 1.55% 516.3000 516.3253 0.0253 0.00%
117 47.8769 49.8816 2.0047 4.19% 67.5198 67.5407 0.0209 0.03% 516.9000 517.9249 1.0249 0.20%
118 47.8769 47.9025 0.0257 0.05% 67.5204 67.5297 0.0093 0.01% 517.9000 518.9306 1.0306 0.20%
119 47.8769 47.8811 0.0042 0.01% 67.5210 67.5444 0.0234 0.03% 516.6000 516.6403 0.0403 0.01%
120 47.8769 48.8822 1.0053 2.10% 67.5216 67.5660 0.0444 0.07% 512.4000 514.4281 2.0281 0.40%
121 47.8768 47.9244 0.0475 0.10% 67.5221 67.5611 0.0389 0.06% 514.2000 514.2438 0.0438 0.01%
122 47.8766 47.9217 0.0450 0.09% 67.5226 67.5420 0.0194 0.03% 513.5000 514.5417 1.0417 0.20%
Average deviation_0.87%_0.45%_0.12%
The Agisoft PhotoScan software capability was evaluated based on a comparison of the data obtained with the Leica TSOO6 total station. As noted earlier, the instrument itself has a level of tolerance (Figure 1), so it is worth paying attention to the smallest and largest values. This problem is extensively discussed in [11], where the object of the study is a specific building with right angles and a flat surface. In our case, the ground, which forms the relief of the object on which the survey was carried out, gives a large error, which is visually reflected in the table above. Minimal error in X-axis was 0.0031, in Y-axis 0.0027 and in Z-axis 0.0031 degrees. Maximum values for X axis was 2.0463, Y axis 1.0477 and Z axis 3.0248 degrees. As expected earlier in the construction of points in the plane difficulties were caused by the triangulation angle, because the scheme of building depends on the geometry of the object, which has no clear reference points in nature. At all this sighting error has a higher priority than instrumental origin and ranges between ±0.3-0.4" in first-class work and ±1" in networks of crowding, which is worth considering when surveying [12]. Similar measurements to confirm the allowable errors were carried out in the article [13] , where the relative error was less than 10%, which corresponds to the real field data. The use of non-metric cameras is quite a serious step for photogravimetry, but correctly chosen software allows to minimize the error range.
A color-coded system was used to indicate the level of difference between the known and obtained coordinates, with green, yellow, and red signifying minimally acceptable, not significant, and maximum critical differences, respectively. The average deviation between the known and Agisoft PhotoScan coordinates was 0.87% in the X-axis, 0.45% in the Y-axis, and 0.12% in the Z-axis. The total station survey took approximately two days to complete for all 122 points, while finding the remaining 121 points only took about three hours due to the process being automated. It should be noted that the time spent working with the software can also be reduced by using a more powerful computer to process the data received from the aircraft.
4. Conclusions
The authors of the article conducted a comparative analysis between traditional and automated processes for determining surveying coordinates. Based on the results of the study, the following conclusions were made:
1. The traditional method allowed for a quick visual estimation of the survey area.
2. The traditional method also allowed for the consideration of terrain peculiarities and existing objects.
3. The automated method provided visualization of the current situation on the construction
site.
4. The automated method allowed for an automatic process of coordinate determination using one initial point obtained with a total station.
5. The study revealed some level of inaccuracy in coordinate values when comparing data obtained from the total station and the software.
References
1. Efficiency optimization and quality control of engineering geodesy processes in civil engineering / V. Berkhahn, F. Berner, H. Kutterer, V. Schwieger, J. Hirschner, I. Rehr, N. Rinke, J. Schweitzer // BAUINGENIEUR. — 2010. — Vol. 85. — P. 491-501.
2. UAS Point Cloud Accuracy Assessment Using Structure from Motion-Based Photogrammetry and PPK Georeferencing Technique for Building Surveying Applications / J.G. Martinez, G. Albeaino, M. Gheisari, W. Volkmann, L.F. Alarcon // Journal of Computing in Civil Engineering. — 2021. — Vol. 35, No. 1. — P. 05020004. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000936
3. Web-Based Architecture for Automating Quantity Surveying Construction Cost Calculation / J.H. Yousif, S. N. Abdul Majeed, F. J. I. Al Azzawi // Infrastructures. — 2020. — Vol. 5, No. 6. — P. 45. https://doi.org/10.3390/infrastructures5060045
4. Geometric accuracy investigations of terrestrial laser scanner systems in the laboratory and in the field / T.P. Kersten, M. Lindstaedt // Applied Geomatics. — 2022. — Vol. 14, No. 2. — P. 421-434. https://doi.org/10.1007/s12518-022-00442-2
5. Assessing the Accuracy of High Resolution Digital Surface Models Computed by PhotoScan® and MicMac® in Sub-Optimal Survey Conditions / M. Jaud, S. Passot, R. Le Bivic, C. Delacourt, P. Grandjean, N. Le Dantec // Remote Sensing. — 2016. — Vol. 8, No. 6. — P. 465. https://doi.org/10.3390/rs8060465
6. Structure from motion processing of analogue images captured by rollei metric camera, digitized with various scanning resolution / A. Dlesk, K. Vach, K. Pavelka // Acta Polytechnica. — 2020. — Vol. 60, No. 4. — P. 288302. https://doi.org/10.14311/AP.2020.60.0288
7. Assessing the performance of commercial Agisoft PhotoScan software to deliver reliable data for accurate3D modelling / A. Jebur, F. Abed, M. Mohammed // MATEC Web of Conferences. — 2018. — Vol. 162. — P. 03022. https://doi.org/10.1051/matecconf/201816203022
8. Performance comparison analysis of 3D reconstruction modeling software in construction site visualization and mapping / A. Keyvanfar, A. Shafaghat, M.S. Rosley // International Journal of Architectural Computing. — 2022. — Vol. 20, No. 2. — P. 453-475. https://doi.org/10.1177/14780771211066876
9. A comparative study between photogrammetry and laser technology applied on model turbine blades / D. Nedelcu, G.R. Gillich, A. Gerocs, I. Padurean // Journal of Physics: Conference Series. — 2020. — Vol. 1426, No. 1. — P. 012026. https://doi.org/10.1088/1742-6596/1426/1/012026
10. The analysis of the accuracy of spatial models using photogrammetric software: Agisoft Photoscan and Pix4D / A. Barbasiewicz, T. Widerski, K. Daliga // E3S Web of Conferences. — 2018. — Vol. 26. — P. 00012. https://doi.org/10.1051/e3sconf/20182600012
11. Assessing the performance of commercial Agisoft PhotoScan software to deliver reliable data for accurate3D modelling [Текст] / A. Jebur, F. Abed, M. Mohammed // MATEC Web of Conferences. — 2018. — Т. 162. — С. 03022. https://doi.org/10.1051/matecconf/201816203022
12. Evaluation of Stereo Matching Algorithms and Dynamic Programming for 3D Triangulation / T.C. Huat, N.A. Manap // Advanced Computer and Communication Engineering Technology: Vol. 315: Lecture Notes in Electrical Engineering. — Cham: Springer International Publishing, 2015. — P. 641-650. https://doi.org/10.1007/978-3-319-07674-4 60
13. Construction and Accuracy Test of a 3D Model of Non-Metric Camera Images Using Agisoft PhotoScan / X. quan Li, Z. an Chen, L. ting Zhang, D. Jia // Procedia Environmental Sciences. — 2016. — Vol. 36. — P. 184-190. https://doi.org/10.1016/j.proenv.2016.09.031
Information about authors:
Zhassulan Kuzbakhov - MSc Student, Department of Civil Engineering, L.N. Gumilyov Eurasian National University, Astana, Kazakhstan, zhaaas0613@gmail.com
Shyngys Zharassov - PhD Student, Department of Civil Engineering, L.N. Gumilyov Eurasian National University, Astana, Kazakhstan, zhshzh95@gmail.com
Author Contributions:
Zhassulan Kuzbakhov - concept, methodology, resources, analysis, editing, funding acquisition. Shyngys Zharassov - data collection, modeling, testing, visualization, interpretation, drafting.
Received: 02.02.2023 Revised: 21.02.2023 Accepted: 21.02.2023 Published: 21.02.2023
