Abstract
Image detection and recognition play an important factor in processing and field of view. In recent years, the rapid growth of deep learning in self-learning features has spanned traditional research algorithms in the field of image research and recognition. Of course, the ability to achieve this is determined by the technical characteristics of deep learning. This is a recently popular technology. Therefore, its changes naturally affect the heartstrings of every researcher. Of course, it can also change the traditional target detection and recognition methods. In order to deeply study the current situation of image detection and recognition algorithms based on deep learning and fractal feature fusion technology, this paper uses graph analysis, recognition detection, and model building methods to collect samples, analyzes image recognition technology, and streamlines the algorithm, and created an image recognition algorithm based on deep learning theory. When studying the recognition rate of the two algorithm models under SOC acquisition conditions, the research results show that the recognition rate of the scn-1 algorithm model can reach 96.12%, and the recognition rate of the Scn-2 algorithm model is 93.55%, which is not only slow in convergence, but also the overall recognition performance is also worse than that of scn-1. In the process of further optimization of the studied algorithm model, the optimization methods sdg and adadelta are compared, and the recognition rate after sdg is finally stable is 92.89% on average. The performance of the adadelta algorithm model is stable when the epoch is 36, and the recognition rate after stability is 95.69% on average. Then set the learning rate attenuation coefficient a = 0.94, and the recognition rate is 96.12%, which shows that the optimization algorithm has good robustness for parameter selection. It is basically realized that starting from deep learning and fractal feature fusion technology, an algorithm model that can accurately and comprehensively detect and recognize images is designed.
Similar content being viewed by others
Data availability
No data were used to support this study.
References
Al-Waisy AS, Qahwaji R, Ipson S et al (2018) A multimodal deep learning framework using local feature representations for face recognition. Mach vis Appl 29(1):35–54
Ambiga N, Nagarajan A (2019) A survey of the elements in the human body and consideration of human bio monitoring for biometric authentication. J Int Pharm Res 46(1):16–19
Bazargani H, Bilaniuk O, Laganiere R (2018) A fast and robust homography scheme for real-time planar target detection. J Real Time Image Proc 15(4):739–758
Chen X, Yang T, Xu J (2016) Multi-gait identification based on multilinear analysis and multi-target tracking. Multimedia Tools Appl 75(11):6505–6532
Chen X, Zhu XB, Chang-Fan WU et al (2021) Research on fundus image classification based on transfer learning and feature fusion. Guangxue Jingmi Gongcheng Opt Precis Eng 29(2):388–399
Ganguly B, Ghosal A, Das A et al (2020) Automated detection and classification of arrhythmia from ECG signals using feature-induced long short-term memory network. IEEE Sens Lett 4(8):1–4
Gu X, Song H, Chen J et al (2020) A review of research on pig behavior recognition based on image processing. Int Core J Eng 6(1):249–254
Guo W, Jiao Z (2018) Tank target image retrieval method based on deep learning and feature fusion. Q J Indian Pulp Pap Tech Assoc 30(7):323–329
Hong Y, Kim J (2019) Art painting detection and identification based on deep learning and image local features. Multimedia Tools Appl 78(6):6513–6528
Jiang M (2020) Research on athlete training behavior based on improved support vector algorithm and target image detection. J Intell Fuzzy Syst 39(4):5725–5736
Konishi Y, Shigematsu K, Tsubouchi T et al (2018) Detection of target persons using deep learning and training data generation for Tsukuba challenge. J Robot Mechatron 30(4):513–522
Kurek J, Swiderski B, Osowski S et al (2018) Deep learning versus classical neural approach to mammogram recognition. Bull Pol Acad Sci Tech Sci 66(6):831–840
Li X, Lei S, Wang Y et al (2018) Fast aerial UAV detection based on image segmentation and HOG-FLD feature fusion. Int J Adv Netw Monit Controls 3(1):106–114
Liu ZG, Yang Y, Ji XH (2016) Flame detection algorithm based on a saliency detection technique and the uniform local binary pattern in the YCbCr color space. Signal Image Video Process 10(2):277–284
Lu L, Wang J (2016) Image processing and recognition algorithm for target tracking. Int J Smart Sens Intell Syst 9(1):353–376
Molina-Molina EO, Solorza-Calderón S, Lvarez-Borrego J (2020) Classification of dermoscopy skin lesion color-images using fractal-deep learning features. Appl Sci 10(5954):1–19
Ren X (2021) Research on a software architecture of speech recognition and detection based on interactive reconstruction model. Int J Speech Technol 24(4):1–9
Sun YK, Chang HK, Cho JH et al (2020) Multiple model guided image filter for target detection and recognition according to noise type. J Inst Control 26(8):616–623
Tang Y, Wu X (2018) Scene text detection using superpixel-based stroke feature transform and deep learning based region classification. Multimedia IEEE Trans 20(9):2276–2288
Wang Y, Liu X, Yu C (2021) Assisted diagnosis of Alzheimer’s disease based on deep learning and multimodal feature fusion. Complexity 2021(4):1–10
Yamane T, Chun PJ (2020) Crack detection from a concrete surface image based on semantic segmentation using deep learning. J Adv Concr Technol 18(9):493–504
Yue Q, Ma C (2016) Study on the deep neural network of intelligent image detection and the improvement of elastic momentum on image recognition. J Comput Theor Nanosci 13(5):3326–3330
Zhang X, Wang Z (2018) Microcalcification clusters detection method based on deep learning and multi-scale feature fusion. Moshi Shibie Yu Rengong Zhineng Pattern Recognit Artif Intell 31(11):1028–1039
Zhang A, He J, Lin Y et al (2020) Recognition of partial discharge of cable accessories based on convolutional neural network with small data set. Int J Comput Math Electr Electron Eng 39(2):431–446
Zheng B, Yun D, Liang Y (2020) Research on behavior recognition based on feature fusion of automatic coder and recurrent neural network. J Intell Fuzzy Syst 39(6):8927–8935
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The author states that this article has 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
Liu, D., Teng, W. Deep learning-based image target detection and recognition of fractal feature fusion for BIOmetric authentication and monitoring. Netw Model Anal Health Inform Bioinforma 11, 17 (2022). https://doi.org/10.1007/s13721-022-00355-5
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s13721-022-00355-5