基于成像声呐的鱼类长度测量误差与修正模型研究

彭战飞; 沈蔚; 张进

doi:10.12131/20220279

基于成像声呐的鱼类长度测量误差与修正模型研究

doi: 10.12131/20220279

上海海洋大学海洋科学学院，上海 201306

基金项目: 2020年上海海洋大学科技发展专项基金；上海城投原水有限公司项目 (D-8006-19-0088)

详细信息

作者简介:
彭战飞 (1993—)，男，助教，硕士，研究方向为鱼类目标声学提取。E-mail: zfpeng@shou.edu.cn

中图分类号: S 923.4
计量
- 文章访问数: 55
- 被引次数: 0
出版历程
- 收稿日期: 2022-10-21
- 修回日期: 2023-03-05
- 录用日期: 2023-03-31
- 网络出版日期: 2023-04-28

Study on error and correction model of fish length measurement based on imaging sonar

College of Marine Sciences, Shanghai Ocean University, Shanghai 201306, China

摘要

摘要: 成像声呐能在扫描波束范围内进行鱼类映像长度的测定，为提高成像声呐测定鱼类长度的精度，文章利用自适应分辨率成像声呐 (Adaptive Resolution Imaging Sonar, ARIS) 开展鱼类长度测定实验，依据ARIS扫描波束内鱼类的映像长度和实测长度，分析了ARIS测量鱼类长度的误差。实验结果表明，鱼类与声呐波束的夹角是导致鱼类长度测量误差的主要因素，夹角越大，鱼类映像长度测量误差越小。在4 m视野内ARIS探测距离对鱼类长度的测量误差无显著交互影响。鱼类映像长度与尾叉长误差更小，其误差均值为2.1 cm。鱼类映像长度与全长、尾叉长的线性关系良好，其线性修正模型拟合度R²分别为0.995 1和0.990 5。研究表明，鱼类映像长度使用尾叉长的测量更为有效，同时鱼类与声呐波束的夹角对映像长度的误差分析和声呐映像长度修正模型，可为获取更准确的鱼类长度信息提供参考。
- 鱼类长度测量 /
- 成像声呐 /
- 测量误差 /
- 修正模型
Abstract: Imaging sonar can measure the length of fish images within its scanning beam. To improve the accuracy of imaging sonar’s measurement of fish length, we conducted an experiment to determine the fish length by using Adaptive Resolution Imaging Sonar (ARIS). We compared and analyzed the fish length error in acoustic image based on the measured length of a trail fish and the image length at a known position of the ARIS scanning beam. The results indicate that the main influencing factor of fish image length measurement error was the angle between the fish and the sonar beam. As the angle increased, the measurement error of fish length decreased. There was no significant interactive effect of detection distance of ARIS on length measurement error of fish in the 4 M field. The error between the fish image length and the fork length was minimal, with an average error of 2.1 cm. Furthermore, the image length of fish had a good linear relationship with the total length and fork length, and the fitting degree R² of the linear modified model was 0.995 1 and 0.990 5, respectively. The study shows that fish image length based on fork measurements is more effective. Additionally, the error analysis of the angle between fish and sonar beam on image length and the sonar image length modified model also provide preliminary references for obtaining more accurate fish length information.
- Measurement of fish length /
- Imaging sonar /
- Measurement error /
- Modified model

图 1 声学实验水池

Figure 1. Acoustic experimental pool

下载: 全尺寸图片幻灯片

图 2 研究思路

Figure 2. Research idea

下载: 全尺寸图片幻灯片

图 3 鱼类固定装置

Figure 3. Fixed device of fish

下载: 全尺寸图片幻灯片

图 4 不同角度实验鱼映像图

注：a. φ 为0°时实验鱼映像图；b. φ 为51°时实验鱼映像图；c. φ为90°时实验鱼映像图。

Figure 4. Images of trial fish from different angles

Note: a. Trial fish image at 0°; b. Trial fish image at 51°; c. Trial fish image at 90°.

下载: 全尺寸图片幻灯片

图 5 ARIS映像测量坐标系

Figure 5. ARIS image measurement system

下载: 全尺寸图片幻灯片

图 6 鱼类目标成像声呐测定误差受探测距离和已知长度的影响。

注：正误差值表示对鱼长高估，而负值表示低估。

Figure 6. Measurement error of imaging sonar of fish target as affected by detection distance and known length

Note: Positive error values represent overestimated fish lengths, while negative values represent underestimated fish lengths.

下载: 全尺寸图片幻灯片

图 7 分别使用传感器的两个频率在水平面展开的侧面测定所有鱼类映像长度

Figure 7. All fish image lengths measured on side of horizontal expansion using two frequencies of sensor

下载: 全尺寸图片幻灯片

图 8 映像长度误差的模型取决于夹角 φ 大小

注：a. 高频鱼类映像长度与全长误差；b. 低频鱼类映像长度与全长误差；c. 高频鱼类映像长度与尾叉长误差；d. 低频鱼类映像长度与尾叉长误差。

Figure 8. Model of image length error depends on size of angle

Note: a. Error between the length and the total length of the high-frequency fish image; b. Error between image length and total length of low-frequency fishimage; c. Error between the length of high-frequency fish image and the length of the tail fork; d. Error between image length and fork length of low-frequency fish.

下载: 全尺寸图片幻灯片

图 9 映像长度误差率

Figure 9. Error rate of image length

下载: 全尺寸图片幻灯片

图 10 鱼类映像长度与全长和尾叉长的线性模型

Figure 10. Linear model of fish image length, total length and fork length

下载: 全尺寸图片幻灯片

表 1 实验鱼长度统计

Table 1. Length statistics of trial fish

编号No.	种类Species	全长 Total length/cm	尾叉长Fork length/cm
1	鲫	47.5	41.8
2	鲫	38.0	35.5
3	鳙	20.5	17.0
4	鲫	26.5	23.5
5	鲫	37.5	34.0
6	鳙	42.0	39.0
7	鳙	44.0	39.0
8	鲫	37.5	34.0

下载: 导出CSV

表 2 映像长度统计

Table 2. Image length statistics

编号No.	映像长度均值Average image length/cm		映像长度误差Image length error/cm		中误差Mean square error/cm
编号No.	1.8 MHz	1.1 MHz	1.8 MHz	1.1 MHz	1.8 MHz	1.1 MHz
1	42.1	40.75	5.4	6.4	2.79	1.8
2	34.35	31.2	3.65	7.8	2.43	1.31
3	18.16	16.55	2.34	3.95	1.98	1.54
4	22.58	22.42	3.95	4.18	1.74	2.34

下载: 导出CSV

参考文献(34)

[1]	SIMMONDS J E, MACLENNAN D N. Fisheries acoustics: theory and practice: 2nd ed.[J]. Fish Fish, 2006, 7: 227-228. doi: 10.1111/j.1467-2979.2006.00220.x
[2]	TUSER M, FROUZOVA J, BALK H, et al. Evaluation of potential bias in observing fish with a DIDSON acoustic camera[J]. Fish Res, 2014, 155: 114-121. doi: 10.1016/j.fishres.2014.02.031
[3]	BELCHE E O, LYNN D C. Acoustic, near-video-quality image for work in turbid water[C]. The Proceedings of Underwater Intervention 2000 Conference, January 2000, Houston, Tex. Houston, Tex: Doyle Publishing Co. Inc, 2000.
[4]	BELCHER E, LYNN D, DINH H, et al. Beamforming and imaging with acoustic lenses in small, high-frequency sonars[C]//Proceedings OCEANS'99 MTS/IEEE Conference & Exhibition, 13–16 September 1999, Seattle, Washington. IEEE Oceanic Engineering Society, 1999, 3: 1495-1499.
[5]	ELLOTT J M, FLETCHER J M. A comparison of three methods for assessing the abundance of Arctic char, Salvelinus alpines Windermere (northwest England)[J]. Fish Res, 2001, 53(1): 39-46. doi: 10.1016/S0165-7836(00)00270-8
[6]	连玉喜, 黄耿, MALGORZATA G, 等. 基于水声学探测的香溪河鱼类资源时空分布特征评估[J]. 水生生物学报, 2015, 39(5): 920-929. doi: 10.7541/2015.121
[7]	刘慧杰, 王从锋, 刘德富, 等. 双频识别声呐在鱼类资源调查中的应用进展[J]. 三峡大学学报(自然科学版), 2015, 37(3): 28-34. doi: 10.13393/j.cnki.issn.1672-948X.2015.03.007
[8]	JECH M J, MICHAELS W L. A multifrequency method to classify and evaluate fisheries acoustic data[J]. Can J Fish Aquat, 2006, 63: 2225-2235. doi: 10.1139/f06-126
[9]	MAXWELL S L, GOVE N E. Assessing a dual-frequency identification sonars' fish-counting accuracy, precision, and turbid river range capability[J]. J Acoust Soc Am, 2007, 122(6): 3364-3377. doi: 10.1121/1.2799500
[10]	LUBIS M Z, MANIK H M. Review: acoustic systems (split beam echo sounder) to determine abundance of fish in marine fisheries[J]. J Geosci Eng Technol, 2017, 2(1): 76-83. doi: 10.24273/jgeet.2017.2.1.38
[11]	ZWOLINSKI J Z, FERNANDES P G F G, MARQUES V, et al. Estimating fish abundance from acoustic surveys: calculating variance due to acoustic backscatter and length distribution error[J]. Can J Fish Aquat, 2009, 66(12): 2081-2095. doi: 10.1139/F09-138
[12]	MOURSUND R A, CARLSON T J, PETERS R D. A fisheries application of a dual-frequency identification sonar acoustic camera[J]. ICES J Mar Sci, 2003, 60(3): 678-683. doi: 10.1016/S1054-3139(03)00036-5
[13]	LOZORI J D, MICHAELS B C. Sonar estimation of salmon passage in the Yukon River near pilot station, 2009[R]. Fairbanks: Alaska Department of Fish and Game, Division of Commercial Fisheries, 2013: 1-3.
[14]	JOHNSON E L, CLABOUGH T S, KEEFER M L, et al. Evaluation of dual frequency identification sonar (DIDSON) for monitoring Pacific Lamprey passage behavior at fishways of Bonneville Dam, 2011[R]. U. S. Army Corps of Engineers Portland District, Idaho Cooperative Fish and Wildlife Research Unit, 2012.
[15]	荆丹翔, 周晗昀, 韩军, 等. 基于成像声呐DIDSON的水域内鱼群数量估计方法[J]. 应用声学, 2019, 38(4): 705-711. doi: 10.11684/j.issn.1000-310X.2019.04.030
[16]	HOLMES J A, CRONKITE G, ENZENHOFER H J, et al. Accuracy and precision of fish-count data from a 'dual-frequency identification sonar' (DIDSON) imaging system[J]. ICES J Mar Sci, 2006, 63(3): 543-555. doi: 10.1016/j.icesjms.2005.08.015
[17]	LIN D Q, ZHANG H, KANG M, et al. Measuring fish length and assessing behavior in a high-biodiversity reach of the Upper Yangtze River using an acoustic camera and echo sounder[J]. J Appl Ichthyol, 2016, 32(6): 1072-7079. doi: 10.1111/jai.13134
[18]	JING D X, HAN J, WANG X D, et al. A method to estimate the abundance of fish based on dual-frequency identification sonar (DIDSON) imaging[J]. Fish Sci, 2017, 83(5): 685-697. doi: 10.1007/s12562-017-1111-3
[19]	张翔, 沈蔚, 童剑锋, 等. 基于DIDSON双频识别声呐技术的青草沙水库鱼类资源量评估[J]. 上海海洋大学学报, 2017, 26(4): 561-569. doi: 10.12024/jsou.20161201919
[20]	徐兆礼, 沈盎绿. 兴化湾海域鱼类多样性的时空变化[J]. 中国水产科学, 2011, 19(2): 174-184.
[21]	CRONKITE G M W. Use of high-frequency imaging sonar to estimate adult sockeye salmon escapement in the Horsefly River, British Columbia[R]. Nanaimo, British Columbia: Fisheries and Oceans Canada, 2006.
[22]	BAUMGARTNER L J, REYNOLDSON N, CAMERON L, et al. Application of a Dual-frequency Identification Sonar (DIDSON) to fish migration studies[C]//Cutting-edge Technologies in Fish and Fisheries Science, 28-29 August, 2006, Hobart, Tasmania, Australia, Australian. Australian Society for Fish Biology, 2006, 84: 1-33.
[23]	CROSSMAN J A, MARTEL G, JOHNSON P N, et al. The use of dual frequency identification sonar (DIDSON) to document white sturgeon activity in the Columbia River, Canada[J]. J Appl Ichthyol, 2011, 27: 53-57.
[24]	NEUMANN R M, GUY C S, WILLIS D W. Length, weight, and associated structural indices[C]//ZALE A V, PARRISH D L, SUTTON T M. Fisheries techniques. Bethesda, Maryland: American Fisheries Society, 2012: 637-676.
[25]	SHIN Y J, ROCHET M J, JENNINGS S, et al. Using size-based indicators to evaluate the ecosystem effects of fishing[J]. ICES J Mar Sci, 2005, 62(3): 384-396. doi: 10.1016/j.icesjms.2005.01.004
[26]	BECKER A, WHITFIELD A K, COWLEY, et al. An assessment of the size structure, distribution and behavior of fish populations within a temporarily closed estuary using dual frequency identification sonar (DIDSON)[J]. J Fish Biol, 2011, 79(3): 761-775. doi: 10.1111/j.1095-8649.2011.03057.x
[27]	SCHRECK C B, MOYLE P B. Methods for fish biology[M]. Bethesda: American Fisheries Society, 1990.
[28]	HIGHTOWER J E, MAGOWAN K J, BROWN L M, et al. Reliability of fish size estimates obtained from a multibeam imaging sonar[J]. J Fish Wildl Manag, 2013, 4(1): 86-96. doi: 10.3996/102011-JFWM-061
[29]	COOK D, MIDDLEMISS K, JAKSONS P, et al. Validation of fish length estimations from a high frequency multi-beam sonar (ARIS)and its utilization as a field-based measurement technique[J]. Fish Res, 2019, 218: 59-68. doi: 10.1016/j.fishres.2019.05.004
[30]	DAROUX A, MARTIGNAC F, NEVOUX M, et al. Manual fish length measurement accuracy for adult river fish using an acoustic camera (DIDSON)[J]. J Fish Biol, 2019, 95(2): 480-489. doi: 10.1111/jfb.13996
[31]	BURWEN D L, FLEISCHMAN S J, MILLER J D. Accuracy and precision of salmon length estimates taken from DIDSON sonar images[J]. Trans Am Fish Soc, 2010, 139(5): 1306-1314. doi: 10.1577/T09-173.1
[32]	GERLOTTO F, GEORGAKARAKOS S, ERIKSEN P. The application of multibeam sonar technology for quantitative estimates of fish density in shallow water acoustic surveys[J]. Aquat Living Resour, 2000, 13: 385-393. doi: 10.1016/S0990-7440(00)01055-X
[33]	BURWEN D L, NEALSON P A, FLISCHMAN S J, et al. The complexity of narrowband echo envelopes as a function of fish side-aspect angle[J]. ICES J Mar Sci, 2007, 64(5): 1066-1074. doi: 10.1093/icesjms/fsm074
[34]	柯森繁, 陈渴鑫, 罗佳, 等. 鲢顶流游泳速度与摆尾行为相关性分析[J]. 水产学报, 2017, 41(3): 401-406.