Improvement of neural network model topology for object segmentation in digital images based on convolutional neural networks
- Authors: Kulikov A.А.1,2
-
Affiliations:
- MIREA – Russian Technological University
- Financial University under the Government of the Russian Federation
- Issue: Vol 11, No 3 (2024)
- Pages: 57-63
- Section: INFORMATICS AND INFORMATION PROCESSING
- URL: https://journal-vniispk.ru/2313-223X/article/view/285909
- DOI: https://doi.org/10.33693/2313-223X-2024-11-3-57-63
- EDN: https://elibrary.ru/QGYEQJ
- ID: 285909
Cite item
Abstract
Nowadays, convolutional neural networks have demonstrated significant performance gains over traditional machine learning methods for various real-world computational intelligence tasks such as digital image classification. However, to achieve the best accuracy, the network topology should be modeled using different architectures with different number of filters, kernel size, number of layers, etc., which actualizes the problem of developing and justifying appropriate selection methods. Taking into account the above mentioned, the aim of the paper is to justify an approach that will improve the topology of the neural network model for object segmentation in digital images based on convolutional neural networks. The research methods are system analysis, modeling, machine learning and fuzzy logic theory, and decision-making theory. As a result of the analysis, the paper proposes an algorithm to improve the topology of the neural network model based on differential evolution to optimize the accuracy of image segmentation and the training time of the network. Differential evolution is applied to determine the optimal number of layers in the network topology, which promotes faster convergence. Within the proposed algorithm, an encoding step was identified to represent the structure of each network using a fixed-length integer array, after which it is proposed to utilize differential evolution processes (mutation, recombination, and selection) to efficiently explore the search space. Prospects for further research are to develop methods and techniques to encode a candidate solution using different numbers of hidden blocks in each convolution.
Keywords
Full Text
##article.viewOnOriginalSite##About the authors
Alexander А. Kulikov
MIREA – Russian Technological University; Financial University under the Government of the Russian Federation
Author for correspondence.
Email: tibult41@gmail.com
ORCID iD: 0000-0002-8443-3684
SPIN-code: 6421-0999
Scopus Author ID: 1095062
Cand. Sci. (Eng.), associate professor, associate professor, Department of Data Analysis and Machine Learning
Russian Federation, Moscow; MoscowReferences
- Tsygulev K.S. Application of convolutional neural networks for object segmentation in images. Scientific, Technical and Economic Cooperation of Asia-Pacific Countries in the 21st Century. 2021. No. 2. Pp. 308–312. (In Rus.)
- Bratukhin D. Segmentation of flames in an image using a convolutional neural network U-NET. Norwegian Journal of Development of the International Science. 2021. No. 64. Pp. 31–33. (In Rus.)
- Godunov A.I., Balanyan S.T., Egorov P.S. Image segmentation and object recognition based on convolutional neural network technology. Reliability and Quality of Complex Systems. 2021. No. 3 (35). Pp. 62–73. (In Rus.)
- Lutsenko V.S., Shukhman A.E. Medical image segmentation using convolutional neural networks. Bulletin of Computer and Information Technologies. 2022. No. 6 (216). Pp. 40–50. (In Rus.)
- Binjun He Wenbin Hu. Image segmentation algorithm of lung cancer based on neural network model. Expert Systems. 2021. No. 39. Pp. 145–152.
- Srinitya G., Sharmila D. Certain investigations on image segmentation algorithms on synthetic aperture radar images and classification using convolution neural network. Concurrency and Computation: Practice and Experience. 2021. No. 34. Pp. 73–79.
- Kai Su, Xin Zhang. Convolutional neural network based image segmentation algorithm for dual-layer LCDs. SID Symposium Digest of Technical Papers. 2022. No. 53. Pp. 110–119.
- Berezovsky I.I. A review of convolutional neural networks for medical image segmentation. Scientist’s Tribune. 2022. No. 6. Pp. 59–67. (In Rus.)
- Lozhkin I.A. Augmentation of image sets for training neural networks when solving semantic segmentation problems. International Journal of Open Information Technologies. 2023. No. 1. Pp. 109–117. (In Rus.)
- Xiao Qing Zhang, Guang Yu Wang. COVSeg-NET: A deep convolution neural network for COVID-19 lung CT image segmentation. International Journal of Imaging Systems and Technology. 2021. No. 31. Pp. 38–46.
- Aarthi Sundaram, Chitrakala Sakthivel. Object detection and estimation: A hybrid image segmentation technique using convolutional neural network model. Concurrency and Computation: Practice and Experience. 2022. No. 34. Pp. 65–73.
- Yuma Hakumura, Taiyo Ito. Loss function for ambiguous boundaries for deep neural network (DNN) for image segmentation. Electronics and Communications in Japan. 2023. No. 4. Pp. 149–153.
- Mikhailov A.A. Automatic data labeling for document image segmentation using deep neural networks. Proceedings of the Institute of System Programming of the Russian Academy of Sciences. 2022. No. 6. Pp. 137–146. (In Rus.)
- Kataev M.Yu., Kartashov E.Yu., Ryabukhin V.V. et al. A technique for segmenting images of unmanned aerial vehicles using neural networks. Modern Problems of Remote Sensing of the Earth from Space. 2023. No. 1. Pp. 55–66. (In Rus.)
- Jie Yang, Yong Chen. Convolutional neural network based on the fusion of image classification and segmentation module for weed detection in alfalfa. Pest Management Science. 2024. No. 56. Pp. 91–99.
- Tayebeh Lotfi Mahyari, Richard M. Dansereau multi-layer random walker image segmentation for overlapped cervical cells using probabilistic deep learning methods. IET Image Processing. 2022. No. 16. Pp. 205–211.
- Chernenky I.M. Segmentation of renal structures from contrast-enhanced computed tomography images using convolutional neural network. Sechenovsky Bulletin. 2023. No. 1. Pp. 39–49. (In Rus.)
- Ginni Arora, Ashwani Kumar. Dubey Architecture of an effective convolutional deep neural network for segmentation of skin lesions in dermoscopic images. Expert Systems. 2021. No. 40. Pp. 69–94. (In Rus.)
- Andrianova A.I., Pereverzeva V.I., Shananin V.A. Application of semantic neural networks in image processing. Innovation and Investment. 2022. No. 11. Pp. 224–227. (In Rus.)
- Michael Osadebey, Marius Pedersen. Enhancement of clustering techniques by coupling clustering tree and neural network: Application to brain tumour segmentation. Expert Systems. 2022. No. 40. Pp. 72–78.
Supplementary files
