Counting, locating, and sizing of shrimp larvae based on density map regression
Tóm tắt
The count, location, and size of shrimp larvae are important observations for their intensive breeding, which could evaluate the culture density, growth, and yield. Recent works on the counting of dense aquatic objects in an image are mainly accomplished by the density map regression based on convolutional neural network (CNN). These regression methods only give the estimated density map and the corresponding count, but cannot locate and size the objects accurately. To realize the noninvasive counting, locating, and sizing during aquatic breeding, we propose a novel detection architecture in computer vision for the larvae of Cherax quadricarinatus. In particular, a new shrimp larvae counting network (SLCNet) using the density map regression is designed to output the estimated density map of the shrimp larvae image and predict the corresponding number of the shrimp larvae. The local peak filtering (LPF) is applied to the estimated density map to locate the coordinates of every shrimp larva. We then use the K-nearest neighbor (KNN) and the bounding box search to find the fitting bounding boxes of the shrimp larvae in sparse regions and calculate the body length of the shrimp larvae in the fitting bounding boxes with the aid of the least squares regression method. Compared with typical CNN-based counting models, the mean absolute error and root mean square error of the proposed SLCNet are 4.13 and 5.75, respectively, and the counting accuracy is up to 98.57%. Then, we perform the detection of the location and size of the shrimp larvae using the density map estimated by the SLCNet. The locating and sizing accuracy can reach 90.6% (i.e., F1 value) and 90.28%, respectively. It proves that this detection architecture could be used as technical support for subsequent studies on larva growth and behavior recognition.
Từ khóa
Tài liệu tham khảo
AlFaris NA, Alshammari GM, AlTamimi JZ, AlMousa LA, Alagal RI, Alkehayez NM, Aljabryn DH, Alsayadi MM, Yahya MA (2022) Evaluating the effects of different processing methods on the nutritional composition of shrimp and the antioxidant activity of shrimp powder. Saudi J Biol Sci 29(1):640–649. https://doi.org/10.1016/j.sjbs.2021.09.029
Boles WW, Geva S, Busch A (1999) An image processing approach for estimating the number of live prawn larvae in water. In: ISSPA'99. Proceedings of the Fifth International Symposium on Signal Processing and its Applications, pp 571–574. https://doi.org/10.1109/ISSPA.1999.815737
Cao X, Wang Z, Zhao Y, Su F (2018) Scale aggregation network for accurate and efficient crowd counting. In: Proceedings of the European Conference on Computer Vision, pp 757–773. https://doi.org/10.1007/978-3-030-01228-1_45
Chen Y, Liu H, Yang L, Yu H, Li D, Mei S, Liu Y (2023) A lightweight detection method for the spatial distribution of underwater fish school quantification in intensive aquaculture. Aquacult Int 31(1):31–52. https://doi.org/10.1007/s10499-022-00963-y
Costa CS, Tetila EC, Astolf G, Sant’Ana DA, Brito Pache MC, Gonçalves AB, Garcia Zanoni VA, PicoliNucci HH, Diemer O, Pistori H (2019) A computer vision system for counting oocytes using images captured by smartphone. Aquacult Eng 87:102017. https://doi.org/10.1016/j.aquaeng.2019.102017
Duan Y, Stien LH, Thorsen A, Karlsen Ø, Sandlund N, Li D, Fu Z, Meier S (2015) An automatic counting system for transparent pelagic fish eggs based on computer vision. Aquacult Eng 67:8–13. https://doi.org/10.1016/j.aquaeng.2015.05.001
Emerenciano MGC, Rombenso AN, Vieira FDN, Martins MA, Coman GJ, Truong HH, Noble TH, Simon CJ (2022) Intensification of Penaeid shrimp culture: an applied review of advances in production systems, nutrition and breeding. Animals 12(3):236. https://doi.org/10.3390/ani12030236
Falk T, Mai D, Bensch R, Çiçek Ö, Abdulkadir A, Marrakchi Y, Böhm A, Deubner J, Jäckel Z, Seiwald K, Dovzhenko A, Tietz O, Dal Bosco C, Walsh S, Saltukoglu D, Tay TL, Prinz M, Palme K, Simons M, Diester I, Brox T, Ronneberger O (2019) U-Net: deep learning for cell counting, detection, and morphometry. Nat Methods 16:67–70. https://doi.org/10.1038/s41592-018-0261-2
FAO (2022) The state of world fisheries and aquaculture 2022. Towards blue transformation, Rome, FAO. https://www.fao.org/3/cc0461en/cc0461en.pdf. Accessed 21 Oct 2023
França Albuquerque PL, Garcia V, da Silva OA, Lewandowski T, Detweiler C, Gonçalves AB, Pistori H (2019) Automatic live fingerlings counting using computer vision. Comput Electron Agric 167:105015. https://doi.org/10.1016/j.compag.2019.105015
Girshick R, Donahue J, Darrell T, Malik J (2014) Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 580–587. https://doi.org/10.1109/CVPR.2014.81
Hou S, Liu J, Wang Y, An D, Wei Y (2022) Research on fish bait particles counting model based on improved MCNN. Comput Electron Agric 196:106858. https://doi.org/10.1016/j.compag.2022.106858
Idrees H, Tayyab M, Athrey K, Zhang D, Al-Maadeed S, Rajpoot N, Shah M (2018) Composition loss for counting, density map estimation and localization in dense crowds. In: Proceedings of the European conference on computer vision/ECCV, pp 532–546. https://doi.org/10.1007/978-3-030-01216-8_33
Juarez LM, Luxem AH, Rouse DB (1996) Sampling shrimp populations in hatcheries. J World Aquacult Soc 27(2):218–222. https://doi.org/10.1111/j.1749-7345.1996.tb00273.x
Kaewchote J, Janyong S, Limprasert W (2018) Image recognition method using local binary pattern and the random forest classifier to count post larvae shrimp. Agric Nat Resour 52(4):371–376. https://doi.org/10.1016/j.anres.2018.10.007
Lai Y (2019) A comparison of traditional machine learning and deep learning in image recognition. J Phys: Conf Ser IOP Publ 1314(1):012148. https://doi.org/10.1088/1742-6596/1314/1/012148
Lemarie DP, Weller DA, Theisen DD, Curry Woods L III (1997) Technical notes: evaluation of an electronic fry counter with Striped Bass embryos and larvae. Prog Fish Cult 59(1):71–74. https://doi.org/10.1577/1548-8640(1997)059%3c0071:TNEOAE%3e2.3.CO;2
Lempitsky V, Zisserman A (2010) Learning to count objects in images. In: Advances in neural information processing systems 23, pp 1324–1332. https://papers.nips.cc/paper_files/paper/2010/file/fe73f687e5bc5280214e0486b273a5f9-Paper.pdf. Accessed 21 Oct 2023
Li D, Miao Z, Peng F, Wang L, Hao Y, Wang Z, Chen T, Li H, Zheng Y (2021) Automatic counting methods in aquaculture: a review. J World Aquacult Soc 52(2):269–283. https://doi.org/10.1111/jwas.12745
Li X, Shang M, Qin H, Chen L (2015) Fast accurate fish detection and recognition of underwater images with fast R-CNN. In: OCEANS 2015-MTS/IEEE Washington, pp 1–5. https://doi.org/10.23919/OCEANS.2015.7404464
Li X, Shang M, Hao J, Yang Z (2016) Accelerating fish detection and recognition by sharing CNNs with objectness learning. In: OCEANS 2016-Shanghai, pp 1–5. https://doi.org/10.1109/OCEANSAP.2016.7485476
Li Y, Zhang X, Chen D (2018) CSRNet: dilated convolutional neural networks for understanding the highly congested scenes. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1091–1100. https://doi.org/10.1109/CVPR.2018.00120
Liang D, Xu W, Zhu Y, Zhou Y (2021) Focal inverse distance transform maps for crowd localization and counting in dense crowd. arXiv preprint arXiv:2102.07925. https://doi.org/10.48550/arXiv.2102.07925
Liu Y, Shi M, Zhao Q, Wang X (2019) Point in, box out: beyond counting persons in crowds. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 6469–6478. https://doi.org/10.1109/CVPR.2019.00663
Ni C, Wang D, Vinson R, Holmes M, Tao Y (2019) Automatic inspection machine for maize kernels based on deep convolutional neural networks. Biosys Eng 178:131–144. https://doi.org/10.1016/j.biosystemseng.2018.11.010
Ran X, Li B, Li D, Wang J, Duan Q (2023) Detection of surfacing white shrimp under hypoxia based on improved lightweight YOLOv5 model. Aquacult Int. https://doi.org/10.1007/s10499-023-01149-w
Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 779–788. https://doi.org/10.1109/CVPR.2016.91
Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. In: International Conference on Medical image computing and computer-assisted intervention, pp 234–241. https://doi.org/10.1007/978-3-319-24574-4_28
Rudloe J, Rudloe A (2009) Shrimp: the endless quest for pink gold. FT Press
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv: 1409.1556. https://doi.org/10.48550/arXiv.1409.1556
Solahudin M, Slamet W, Dwi AS (2018) Vaname (Litopenaeus vannamei) shrimp fry counting based on image processing method. IOP Conf Ser: Earth Environ Sci IOP Publ 147(1):012014. https://doi.org/10.1088/1755-1315/147/1/012014
Song Q, Wang C, Jiang Z, Wang Y, Tai Y, Wang C, Li J, Huang F, Wu Y (2021) Rethinking counting and localization in crowds: a purely point-based framework. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 3365–3374. https://doi.org/10.1109/ICCV48922.2021.00335
Thai TTN, Nguyen TS, Pham VC (2021) Computer vision based estimation of shrimp population density and size. In: 2021 International symposium on electrical and electronics engineering, pp 145–148. https://doi.org/10.1109/ISEE51682.2021.9418638
Yada S, Chen H (1997) Weighing type counting system for seedling fry. Bull Jpn Soc Sci Fish (Japan) 63(2):178–183. https://doi.org/10.2331/suisan.63.178
Yang X, Zhang S, Liu J, Gao Q, Dong S, Zhou C (2021) Deep learning for smart fish farming: applications, opportunities and challenges. Rev Aquac 13(1):66–90. https://doi.org/10.1111/raq.12464
Yeh CT, Ling MS (2021) Portable Device for ornamental shrimp counting using unsupervised machine learning. Sens Mater 33(9):3027–3036. https://doi.org/10.18494/SAM.2021.3240
Zhang J, Yang G, Sun L, Zhou C, Zhou X, Li Q, Guo J (2021) Shrimp egg counting with fully convolutional regression network and generative adversarial network. Aquacult Eng 94:102175. https://doi.org/10.1016/j.aquaeng.2021.102175
Zhang L, Zhou X, Li B, Zhang H, Duan Q (2022) Automatic shrimp counting method using local images and lightweight YOLOv4. Biosys Eng 220:39–54. https://doi.org/10.1016/j.biosystemseng.2022.05.011
Zhang Y, Zhou D, Chen S, Gao S, Ma Y (2016) Single-image crowd counting via multi-column convolutional neural network. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 589–597. https://doi.org/10.1109/CVPR.2016.70
Zhao S, Zhang S, Liu J, Wang H, Zhu J, Li D, Zhao R (2021) Application of machine learning in intelligent fish aquaculture: a review. Aquaculture 540:736724. https://doi.org/10.1016/j.aquaculture.2021.736724
Zhao Y, Li W, Li Y, Qi Y, Li Z, Yue J (2022) LFCNet: a lightweight fish counting model based on density map regression. Comput Electron Agric 203:107496. https://doi.org/10.1016/j.compag.2022.107496