Abstract
Authenticity in multimedia information retrieval system is one such parameter which is of utmost in demand. One of such parameters is the biometric based authentication system, as seen in case of human the system proves its importance in almost each and every field. But in case of animals it becomes quite complex to recognize them using unique biometric parameter. This paper presents a modified 10 layered DenseNet deep learning framework which is implemented for the identification of Humpback whale, in an online challenge conducted by the kaggle, using biometric parameter. The proposed dense model minimizes the traditional limitation of vanishing gradient problem and outperforms as compared with other methods exists in the literature. This paper also proposed an hand free flow algorithm for ROI segmentation. In addition, the proposed methodology also uses CLAHE as a preprocessing method and focused on the feature re-usability with an optimized number of parameters used for recognition. During the experimentation, validation of the proposed model is found satisfactory with an accuracy of 93.21%. The test set results are provided by the kaggle with an AUC-ROC result over public and private scoreboard with an accuracy of 94.47% and 92.54% respectively. The comparative result with other deep methodologies suggests that the highest accuracy gained is 95.55% by Se-ResNet50 by some other learderboard and proposed model is found at second place.
Similar content being viewed by others
Notes
All the source images were taken from Google images under the search of animal identification
References
Araabi RN et al (2000) Generalization of dorsal fin ratio for dolphin photo-identification. Ann Biomed Eng 28(10):1269–1279
Araahi RN (2001) Syntactic/semantic curve-matching and photo-identification of dolphins and whales. PhD Thesis, Texas A&M University
Bogucki R et al (2018) Applying deep learning to right whale photoidentification. Conserv Biol 33(3):676–684
Constantine R, et al. (2012) Abundance of humpback whales in oceania using photo-identification and micro satellite genotyping. Mar Ecol Prog Ser 453:249–261
Domeier ML, Nasby-Lucas N (2007) Annual re-sightings of photographically identified white sharks (Carcharodoncarcharias) at an eastern Pacific aggregation site. Mar Biol 150:977–984
Duda J (2019) SGD momentum optimizer with step estimation by online parabola model. pp 1–4. arXiv:1907.07063
Hillman G et al (2003) Computer-assisted photo-identification of flukes using blotch and scar patters. In: Proceedings of 15th biennial conference on the biology of marine mammals
Huang G et al (2017) Densely Connected Convolutional Networks. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 2261–2269
Katona S, et al. (1979). In: Winn HE (ed) Identification of humpback whales by fluke photographs, behavior of marine animals, chapter 2. Plenum Press, New York, pp 33–44
Kehtarnavaz N et al (2003) Photo-identification of humpback and gray whales using affine moment invariants. In: Proceedings of 13th scandinavian conference on image analysis
Kumar S, Singh SK (2017) Visual animal biometrics: survey. IET Biom 6(3):139–156
Leslie NS (2018) A disciplined approach to neural network hyper-parameters: Part 1 – learning rate, batch size, momentum, and weight decay, US Naval Research Laboratory Technical Report 5510-026. arXiv:1803.09820v2
López LFM (2019) Pattern recognition with convolutional neural networks: humpback whale tails. pp 1–16 . Preprints www.preprints.org
Mizroch S, Harkness SAD (2003) A test of computer-assisted matching using the North Pacific humpback whale, Megapteranovaeangliae, tail flukes photograph collection. Mar Fish Rev 65:25–37
Mizroch S et al (1990) Computer assisted photo-identification of humpback whales, individual recognition of cetaceans. International Whaling Commission, Cambridge
Ranguelova E, er al (2004) Towards computer assisted photo identification of humpback whales. In: International conference on image processing (ICIP), pp 1727–1730
Shorten C, Khoshgoftaar TM (2019) A survey on image data augmentation for deep learning. J Big Data 6(60):1–48
Smith LN (2017) Cyclical learning rates for training neural networks. In: IEEE winter conference on applications of computer vision (WACV), pp 464–472
Vidyarthi A, et al. (2019) Classification of breast microscopic imaging using hybrid CLAHE-CNN deep architecture. In: Twelfth international conference on contemporary computing (IC3), pp 1–5
Whale Sense Identifying individualhumpback whales, stellwagen bank national marine sanctuary research and monitoring fact sheet. https://nmsstellwagen.blob.core.windows.net/humpback-id.pdf
WhaleNet. http://whale.wheelock.edu
Zuiderveld K (1994) Contrast limited adaptive histogram equalization. Grap Gems, pp 474–485
Acknowledgements
The author is thankful towards the host organization for providing the basic facilities to implement this project. The author is also thankful towards Google Colab for providing the GPU session and students of the U.G. from host organization in helping to implement this project.
Funding
The author declares that there is no funding associated for this project.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Ethics Statement
The author of this manuscript confirms that: (i) Informed, written consent has been obtained from the relevant sources wherever is required; (ii) All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and its later amendments. (iii) The approval and/or informed consent were not required for study as the dataset is collected from the Kaggle website, a well known data repository for worldwide data analytics problem and live challenges.
Conflict of Interests
The authors of this manuscript declare that there is no conflict of interest.
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
Vidyarthi, A., Malik, A. A hybridized modified densenet deep architecture with CLAHE algorithm for humpback whale identification and recognition. Multimed Tools Appl 81, 19779–19793 (2022). https://doi.org/10.1007/s11042-021-11034-4
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-021-11034-4