Abstract
In cooperation with biologists, we discuss the problem of animal species protection with the usage of modern technologies, namely mobile phones. In our work, we consider the problem of dragonfly image classification, where the aim is given to a preprocessing—segmentation of a dragonfly body from a background. To solve the task, we improve U-Net architecture by ResNeXt cells firstly. Further, we focus on the reasonability of features in neural networks with cardinality dimension and propose the cascaded way of re-using the features among blocks in particular cardinal dimensions. The reuse of the already trained features leads to composing more robust features and more efficient usage of neural network parameters. We test our cascaded cells together with three various U-Net versions for four different settings of hyperparameters with the conclusion that the system of cascaded features leads to higher accuracy than the other versions with the same number of parameters. Also, the cascaded cells are more robust to overfitting the dataset. The obtained results are confirmed on two additional public datasets.
Similar content being viewed by others
Notes
colab.research.google.com/drive/1575oeJWafK9biTGOYmzXMsqdQVoyi50d.
References
Sandra D, Joseph Fargione F (2006) Stuart Chapin III, and David Tilman. Biodiversity loss threatens human well-being. PLoS Biol 4(8):277
Clausnitzer V, Kalkman VJ, Ram M, Collen B, Baillie JEM, Bedjanič M, Darwall WRT, Klaas-Douwe BDijkstra, Rory Dow, John Hawking, et al (2009) Odonata enter the biodiversity crisis debate: the first global assessment of an insect group. Biol Conserv 142(8):1864–1869
ThomasE K, JasonT B (2014) Adult odonata conservatism as an indicator of freshwater wetland condition. Ecol Indicat 38:31–39
Seidu I, Nsor CA, Danquah E, Lancaster L (2018) Odonata assemblages along an anthropogenic disturbance gradient in ghana’s eastern region. Odonatologica
Martín R, Maynou X (2016) Dragonflies (insecta: Odonata) as indicators of habitat quality in mediterranean streams and rivers in the province of barcelona (catalonia, iberian peninsula). Int J Odonatol 19(3):107–124
de Paiva SD, De Marco P, Resende DC (2010) Adult odonate abundance and community assemblage measures as indicators of stream ecological integrity: a case study. Ecol Indicat 10(3):744–752
Kalkman VJ, Clausnitzer V, Dijkstra K-DB, Orr AG, Paulson DR, van Tol J (2007) Global diversity of dragonflies (odonata) in freshwater. In: Freshwater animal diversity assessment. Springer, pp 351–363
Jeno LM, Grytnes J-A, Vandvik V (2017) The effect of a mobile-application tool on biology students’ motivation and achievement in species identification: a self-determination theory perspective. Comput Educ 107:1–12
Ožana S, Burda M, Hykel M, Malina M, Prášek M, Bárta D, Dolnỳ A (2019) Dragonfly hunter cz: mobile application for biological species recognition in citizen science. PloS One 14(1):e0210370
Yeager WC (1932) Some dragonflies of northwest Iowa. In: Proceedings of the Iowa Academy of Science, vol 39, pp 261–263
Long J, Shelhamer Evan D (2015) Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 3431–3440
Zhao H, Qi X, Shen X, Shi J, Jia J (2018) Icnet for real-time semantic segmentation on high-resolution images. In: Proceedings of the European conference on computer vision (ECCV), pp 405–420
Lin T-Y, Dollár P, Girshick R, He K, Hariharan B, Belongie S (2017) Feature pyramid networks for object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2117–2125
Zhao H, Shi J, Qi X, Wang X, Jia J (2017) Pyramid scene parsing network. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2881–2890
Chaurasia A, Culurciello E (2017) Linknet: exploiting encoder representations for efficient semantic segmentation. In: IEEE visual communications and image processing (VCIP). IEEE, pp 1–4
Ross G, Jeff D, Trevor D, Jitendra M (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
Bennett KP, Campbell C (2000) Support vector machines: hype or hallelujah? ACM SIGKDD Explor Newsl 2(2):1–13
Girshick R (2015) Fast R-CNN. In: Proceedings of the IEEE international conference on computer vision, pp 1440–1448
Ren S, He K, Girshick R, Sun J (2015) Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in neural information processing systems, pp 91–99
He K, Gkioxari G, Dollár P, Girshick R (2017) Mask R-CNN. In Proceedings of the IEEE international conference on computer vision, pp 2961–2969
Redmon J, Farhadi A (2018) Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767
Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C-Y, Berg AC (2016) SSD: single shot multibox detector. In: European conference on computer vision. Springer, pp 21–37
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. Springer, pp 234–241
Ning F, Delhomme D, LeCun Y, Piano F, Bottou L, Barbano PE (2005) Toward automatic phenotyping of developing embryos from videos. IEEE Trans Image Process 14:1360–1371
Hinton GE, Krizhevsky A, Wang SD (2011) Transforming auto-encoders. In: International conference on artificial neural networks. Springer, pp 44–51
Scherer D, Müller A, Behnke S (2010) Evaluation of pooling operations in convolutional architectures for object recognition. In: International conference on artificial neural networks. Springer, pp 92–101
Hanin Boris (2018) Which neural net architectures give rise to exploding and vanishing gradients? In: Advances in neural information processing systems, pp 582–591
Vesal S, Ravikumar N, Maier A (2019) A 2d dilated residual u-net for multi-organ segmentation in thoracic ct. arXiv preprint arXiv:1905.07710
Zhang Z, Liu Q, Wang Y (2018) Road extraction by deep residual u-net. IEEE Geosci Rem Sens Lett 15(5):749–753
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778
Xie S, Girshick R, Dollár P, Tu Z, He K (2017) Aggregated residual transformations for deep neural networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1492–1500
He K, Zhang X, Ren S, Sun J (2015) Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Trans Pattern Anal Mach Intell 37(9):1904–1916
Wells WM (1986) Efficient synthesis of gaussian filters by cascaded uniform filters. IEEE Trans Pattern Anal Mach Intell 2:234–239
Canny J (1987) A computational approach to edge detection. In: Readings in computer vision. Elsevier, pp 184–203
Sato Y, Tamura S (1988) Design methods for cascaded gaussian filters. Syst Comput Jpn 19(12):24–34
Hawkins DM (2004) The problem of overfitting. J Chem Inf Comput Sci 44(1):1–12
Zeiler Matthew D (2012) Adadelta: an adaptive learning rate method. arXiv preprint arXiv:1212.5701
Nair V, Hinton GE (2010) Rectified linear units improve restricted Boltzmann machines. In: Proceedings of the 27th international conference on machine learning (ICML-10), pp 807–814
Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167
Wan L, Zeiler M, Zhang S, Cun YL, Fergus R (2013) Regularization of neural networks using dropconnect. In: International conference on machine learning, pp 1058–1066
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE international conference on computer vision, pp 1026–1034
Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the 13th international conference on artificial intelligence and statistics, pp 249–256
Acknowledgements
This research was supported by the project “LQ1602 IT4Innovations excellence in science” and by Student Grant Competition of University of Ostrava (no. SGS06/ UVAFM/2019). For more supplementary materials and overview of our laboratory work, see http://www.graphicwg.irafm.osu.cz/storage/pr/links.html.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Authors Petr Hurtik and Stanislav Ozana declare that they have no conflict of interest.
Human and animal rights
This article does not contain any studies with human participants or animals performed by any of the authors.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Hurtik, P., Ozana, S. Dragonflies segmentation with U-Net based on cascaded ResNeXt cells. Neural Comput & Applic 33, 4567–4578 (2021). https://doi.org/10.1007/s00521-020-05274-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-020-05274-y