Abstract
Traditional underwater image processing methods have low accuracy and slow recognition speed. Underwater multi-target recognition is an important part of multi-target tracking. To solve this problem, an underwater multi-target detection method based on multi-scale convolution neural network (MSC-CNN) is proposed. Firstly, according to the problem of low contrast caused by the absorption and attenuation of underwater light, using the attenuation theory to equalize the dark channel of image to reduce the interference noise of background light in order to enhance the feature signal of object. Secondly, using MSC-CNN to fuse the feature maps generated by different convolution layers in the form of down-sampling and 1*1 convolution to extract the feature. Using the idea of Faster R-CNN for reference, the region-of-interest network is used to select regions of interest and share convolution layers. Finally, Softmax Loss is used to calculate classification probability and non-maximum suppression to correct the border position. The accuracy and test time are taken as evaluation metrics, and the recognition effect of different methods for underwater image is analyzed by comparing with the existing top algorithms. The experimental results of underwater images show that this method has advantages under various evaluation metrics.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
He, B., Wang, G.: Seabed image stitching algorithm with different view points. Microcomput. Inf. 26, 152–154 (2005)
Zhao, H., Xia, S., Zhao, J., Zhu, D., Yao, R., Niu, Q.: Pareto-based many-objective convolutional neural networks. In: Meng, X., Li, R., Wang, K., Niu, B., Wang, X., Zhao, G. (eds.) WISA 2018. LNCS, vol. 11242, pp. 3–14. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-02934-0_1
Ancuti, C., Ancuti, C.O., Haber, T., et al.: Enhancing underwater images and videos by fusion. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 81–88. IEEE (2012)
Chiang, J.Y., Chen, Y.C.: Underwater image enhancement by wavelength compensation and dehazing. IEEE Trans. Image Proc. 21(4), 1756–1769 (2012)
Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1137–1149 (2017)
Zhang, M., Wu, J., Yu, H., et al.: A Novel Fish Image Retrieval Method Based on Saliency Spatial Pyramid. In: 14th International Symposium on Pervasive Systems, Exeter, United Kingdom, 2018, pp. 312–317 (2017)
Sarigül, M., Avci, M.: Comparison of different deep structures for fish classification. In: International Conference on CIT (2017)
Zhuang, P., Xing, L., Liu, Y., et al.: Marine animal detection and recognition with advanced deep learning models. In: Working Notes of CLEF (2017)
Li, X., Shang, M., Qin, H., Chen, L.: Fast accurate fish detection and recognition of underwater images with fast R-CNN. In: OCEANS 2015 MTS/IEEE Washington, Chicago. IEEE (2018)
Lin, W., He, Z.: Application of faster R-CNN model in vehicle detection. Comput. Appl. 38(3), 666–670 (2018)
Garcia, R., Nicosevici, T., Cufi, X.: On the way to solve lighting problems in underwater imaging. In: Oceans, vol. 2, pp. 1018–1024. IEEE (2003)
He, K., Sun, J., Tang, X.: Single image haze removal using dark channel prior. In: IEEE Conference on Computer Vision & Pattern Recognition. IEEE (2009)
Yao, X., Wan, L., Huo, H., et al.: Aircraft target detection in high-resolution remote sensing images based on multi-structure convolutional neural network. Comput. Eng. 43(1), 259–267 (2017)
He, K., Sun, J., Tang, X.: Guided image filtering. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010. LNCS, vol. 6311, pp. 1–14. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-15549-9_1
Lin, T.Y., Dollar, P., Girshick, R., et al.: Feature pyramid networks for object detection, pp. 936–944 (2016)
He, K., Zhang, X., Ren, S., et al.: Deep residual learning for image recognition, pp. 770–778 (2015)
Lécun, Y., Bottou, L., Bengio, Y., et al.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)
Chuang, M.C., Hwang, J.N., Williams, K.: Supervised and unsupervised feature extraction methods for underwater fish species recognition. In: Computer Vision for Analysis of Underwater Imagery, pp. 33–40. IEEE (2014)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)
He, K., Gkioxari, G., Dollar, P., Girshick, R.: Mask r-cnn. arXiv:1703.06870 (2017)
Everingham, M., Van Gool, L., Williams, C.K.I., Winn, J., Zisserman, A.: The pascal visual object classes (VOC) challenge. IJCV 88(2), 303–338 (2010)
Acknowledgment
This research received financial support from the Key R&D Project of Hainan province (Grant No: ZDYD2019020), the National Key R&D Program of China (Grant No: 2018YFB1404401 and 2018YFB1404403), the National Natural Science Foundation of China (Grant No: 61662019 and 61862020), the Education Department of Hainan Province (Grant No: Hnky2019-22), the Higher Education Reform Key Project of Hainan province (Hnjg2017ZD-1) and Academician Workstation in “Hainan” Intelligent Healthcare Technologies.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Huang, S., Huang, M., Zhang, Y., Li, M. (2019). Under Water Object Detection Based on Convolution Neural Network. In: Ni, W., Wang, X., Song, W., Li, Y. (eds) Web Information Systems and Applications. WISA 2019. Lecture Notes in Computer Science(), vol 11817. Springer, Cham. https://doi.org/10.1007/978-3-030-30952-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-30952-7_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-30951-0
Online ISBN: 978-3-030-30952-7
eBook Packages: Computer ScienceComputer Science (R0)