Abstract
In this paper, we have discussed the architecture of certain deep learning algorithms namely, Deconvolutional Neural Network and UNET Network. These are compared to the performances of models based on each of them using various mathematical parameters. Briefly, in both of these architectures, the image is passed through several convolution layers, each followed by a rectified linear unit (ReLU) and a max-pooling operation in the contraction path. This enables to capture the feature information. A similar symmetric expanding path helps find spatial information by passing the image through some up-convolution layers. In UNET though, in the expanding path, the spatial information obtained is concatenated with feature information that was obtained from the contraction path. For the CityScapes Dataset, we can see that models based on UNET clearly outperform the prior models based on Deconvolution Network by evaluating and comparing their IOU values. The network was trained on Google Colab’s GPU.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Wiley, V., Lucas, T.: Computer vision and image processing: a paper review. Int. J. Artif. Intell. Res. 2, 29–36 (2018)
Barik, D., Mondal, M.: Object identification for computer vision using image segmentation. In: 2010 2nd International Conference on Education Technology and Computer, pp. V2-170–V2-172 (2010). https://doi.org/10.1109/ICETC.2010.5529412
Wang, L., Chen, X., Hu, L., Li, H.: Overview of image semantic segmentation technology. In: 2020 IEEE 9th Joint International Information Technology and Artificial Intelligence Conference (ITAIC), pp. 19–26 (2020)
Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., Schiele, B.: The Cityscapes dataset for semantic urban scene understanding. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)
Noh, H., Hong, S., Han, B.: Learning deconvolution network for semantic segmentation. In: 2015 IEEE International Conference on Computer Vision, pp. 1520-1528 (2015)
Zeiler, M.D., Krishnan, D., Taylor, G.W., Fergus, R.: Deconvolutional networks. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 2528–2535 (2010). https://doi.org/10.1109/CVPR.2010.5539957
Albawi, S., Mohammed, T.A., Al-Zawi, S.: Understanding of a convolutional neural network. In: International Conference on Engineering and Technology (ICET), pp. 1–6 (2017). https://doi.org/10.1109/ICEngTechnol.2017.8308186
Chauhan, R., Ghanshala, K.K., Joshi, R.C.: Convolutional neural network (CNN) for image detection and recognition. In: 1st International Conference on Secure Cyber Computing and Communication (ICSCCC), pp. 278–282 (2018). https://doi.org/10.1109/ICSCCC.2018.8703316
“Review: DeconvNet — Unpooling Layer (Semantic Segmentation)”, Towardsdatascience by Sik-Ho Tsang. Referred from https://towardsdatascience.com/review-deconvnet-unpooling-layer-semantic-segmentation-55cf8a6e380e. Accessed 8 Oct 2018
Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: Medical Image Computing and Computer-Assisted Intervention – MICCAI, vol. 9351. Springer (2015)
“UNet — Line by Line Explanation”, Towardsdatascience by Jeremy Zhang. Referred from https://towardsdatascience.com/unet-line-by-line-explanation-9b191c76baf5. Accessed 18 Oct 2019
Rezatofighi, H., Tsoi, N., Gwak, J., Sadeghian, A., Reid, I., Savarese, S.: Generalized intersection over union: a metric and a loss for bounding box regression. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 658–666 (2019). https://doi.org/10.1109/CVPR.2019.00075
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Jayesh Kothari, J., Racha, S.S., Sengupta, J. (2022). Performance of Deconvolution Network and UNET Network for Image Segmentation. In: Bhateja, V., Tang, J., Satapathy, S.C., Peer, P., Das, R. (eds) Evolution in Computational Intelligence. Smart Innovation, Systems and Technologies, vol 267. Springer, Singapore. https://doi.org/10.1007/978-981-16-6616-2_34
Download citation
DOI: https://doi.org/10.1007/978-981-16-6616-2_34
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-6615-5
Online ISBN: 978-981-16-6616-2
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)