Abstract
About fifty percent of the world relies on seafood as main protein source. Due to this, the illegal and uncultured fishery activities are proving to be a threat to marine life. The paper discusses a novel technique that automatically detects and classifies various species of fishes such as dolphins, sharks etc. to help and protect endangered species. Images captured through boat-cameras have various hindrances such as fluctuating degrees of luminous intensity and opacity. The system implemented, aims at helping investigators and nature conservationists, to analyze images of fishes captured by boat-cameras, detect and classify them into species of fishes based on their features. The system adapts to the variations of illumination, brightness etc. for the detection process. The system incorporates a three phase methodology. The first phase is augmentation. This phase involves using data augmentation techniques on real time images dataset captured by boat-cameras and is passed to the detection module. The second phase is the detection. This phase involves detecting fishes in the image by searching for regions in the image having high probability of fish containment. The third phase is the classification of the detected fish into its species. This step involves the segmented image of fishes to be passed to the classifier model which specifies to which species the detected fish belongs to. CNN (Convolutional neural network) is used at the detection and classification phase, with different architectures, to extract and analyze features. The system provides confidence quotients on each image, expressed on a 0–1 scale, indicating the likelihood of the image belonging to each of the following eight categories ALB, BET, YFT, LAG, DOL, Shark, Other and None. The system provides detection and classification with an accuracy of 90% and 92% respectively.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Lowe, D.G.: Distinctive image features from scale-invariant key-points. IJCV 60, 91–110 (2004)
Clara Shanthi, G., Saravanan, E.: Background subtraction techniques: systematic evaluation and comparative analysis. Int. J. Mod. Eng. Res. (IJMER) 3, 514–517 (2013)
Jones, V.: Rapid object detection using a boosted cascade of simple features. In: Computer Vision and Pattern Recognition (2001)
Krizhevsky, A., et al.: ImageNet classification with deep convolutional neural networks. In: NIPS (2012)
He, K., et al.: Deep residual learning for image recognition. ISLR (2015)
Alex, et al.: Dropout: a simple way to avoid overfitting in the network. JMLR (2014)
Szegedy, C., Liu, W., et al.: Going deeper with convolutions. In: CVPR. Google Research (2015)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large scale image recognition. In: ICLR. Visual Geometry Group, Department of Engineering Science, University of Oxford (2015)
Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 580–587 (2014)
Girshick, R.: Fast R-CNN. In: The IEEE International Conference on Computer Vision (ICCV), pp. 1440–1448 (2015)
Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems 28, NIPS (2015)
Kuo, W., Hariharan, B., Malik, J.: DeepBox: learning objectness with convolutional networks. In: The IEEE International Conference on Computer Vision (ICCV), pp. 2479–2487 (2015)
Dai, J., Li, Y., He, K., Sun, J.: R-FCN: object detection via region-based fully convolutional networks. In: Advances in Neural Information Processing Systems 29, NIPS 2016 (2016)
Uijlings, J.R.R., van de Sande, K.E.A., Gevers, T., Smeulders, A.W.M.: Selective search for object recognition. Int. J. Comput. Vis. 104(2), 154–171 (2013)
Lawrence Zitnick, C., Dollr, P.: Edge boxes: locating object proposals from edges. In: European Conference on Computer Vision, ECCV 2014, pp. 391–405 (2014)
Tang, S., Yuan, Y.: Object detection based on convolutional neural network, Stanford Project report (2016)
Hosang, J., Benenson, R.: How good are detection proposals, really? Computer Vision and Pattern Recognition, arXiv (2014)
Gidaris, S., Komodakis, N.: LocNet: improving localization accuracy for object detection, arXiv (2016)
Sermanet, P., Eigen, D., Zhang, X., Mathieu, M., Fergus, R., LeCun, Y.: OverFeat: integrated recognition, localization and detection using convolutional networks, arXiv (2014)
Jaderberg, M., Simonyan, K., Zisserman, A., Kavukcuoglu, K.: Spatial transformer networks, arXiv (2016)
Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: CVPR (2015)
Kingma, D.P., Ba, J.L.: Adam: a method for stochastic optimization. In: ICLR (2015)
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014)
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift, arXiv (2015)
Liu, W., Anguelov, D., Erhan, D., Szegedy, C.: SSD: Single Shot MultiBox Detector, arXiv (2016)
Mnih, V., Heess, N., Graves, A., Kavukcuoglu, K.: Recurrent models of visual attention. In: ICLR. Google DeepMind (2016)
Liang, M., Hu, X.: Recurrent convolutional neural network for object recognition. In: CVPR (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Ethics declarations
✓ All authors declare that there is no conflict of interest.
✓ No humans/animals involved in this research work.
✓ We have used our own data.
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Rekha, B.S., Srinivasan, G.N., Reddy, S.K., Kakwani, D., Bhattad, N. (2020). Fish Detection and Classification Using Convolutional Neural Networks. In: Smys, S., Tavares, J., Balas, V., Iliyasu, A. (eds) Computational Vision and Bio-Inspired Computing. ICCVBIC 2019. Advances in Intelligent Systems and Computing, vol 1108. Springer, Cham. https://doi.org/10.1007/978-3-030-37218-7_128
Download citation
DOI: https://doi.org/10.1007/978-3-030-37218-7_128
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-37217-0
Online ISBN: 978-3-030-37218-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)