Fractional derivative based weighted skip connections for satellite image road segmentation

Authors: Arora, S., Suman, H.K., Mathur, T., Pandey, H.M. and Tiwari, K.

Journal: Neural Networks

Volume: 161

Pages: 142-153

eISSN: 1879-2782

ISSN: 0893-6080

DOI: 10.1016/j.neunet.2023.01.031

Abstract:

Segmentation of a road portion from a satellite image is challenging due to its complex background, occlusion, shadows, clouds, and other optical artifacts. One must combine both local and global cues for an accurate and continuous/connected road network extraction. This paper proposes a model using fractional derivative-based weighted skip connections on a densely connected convolutional neural network for road segmentation. Weights corresponding to the skip connections are determined using Grunwald–Letnikov fractional derivative. Fractional derivatives being non-local in nature incorporates memory into the system and thereby combine both local and global features. Experiments have been performed on two open source widely used benchmark databases viz. Massachusetts Road database (MRD) and Ottawa Road database (ORD). Both these datasets represent different road topography and network structure including varying road widths and complexities. Result reveals that the proposed system demonstrated better performance than the other state-of-the-art methods by achieving an F1-score of 0.748 and the mIoU of 0.787 at fractional order 0.4 on the MRD and a mIoU of 0.9062 at fractional order 0.5 on the ORD.

https://eprints.bournemouth.ac.uk/38089/

Source: Scopus

Fractional derivative based weighted skip connections for satellite image road segmentation.

Authors: Arora, S., Suman, H.K., Mathur, T., Pandey, H.M. and Tiwari, K.

Journal: Neural Netw

Volume: 161

Pages: 142-153

eISSN: 1879-2782

DOI: 10.1016/j.neunet.2023.01.031

Abstract:

Segmentation of a road portion from a satellite image is challenging due to its complex background, occlusion, shadows, clouds, and other optical artifacts. One must combine both local and global cues for an accurate and continuous/connected road network extraction. This paper proposes a model using fractional derivative-based weighted skip connections on a densely connected convolutional neural network for road segmentation. Weights corresponding to the skip connections are determined using Grunwald-Letnikov fractional derivative. Fractional derivatives being non-local in nature incorporates memory into the system and thereby combine both local and global features. Experiments have been performed on two open source widely used benchmark databases viz. Massachusetts Road database (MRD) and Ottawa Road database (ORD). Both these datasets represent different road topography and network structure including varying road widths and complexities. Result reveals that the proposed system demonstrated better performance than the other state-of-the-art methods by achieving an F1-score of 0.748 and the mIoU of 0.787 at fractional order 0.4 on the MRD and a mIoU of 0.9062 at fractional order 0.5 on the ORD.

https://eprints.bournemouth.ac.uk/38089/

Source: PubMed

Fractional derivative based weighted skip connections for satellite image road segmentation

Authors: Arora, S., Suman, H.K., Mathur, T., Pandey, H.M. and Tiwari, K.

Journal: NEURAL NETWORKS

Volume: 161

Pages: 142-153

eISSN: 1879-2782

ISSN: 0893-6080

DOI: 10.1016/j.neunet.2023.01.031

https://eprints.bournemouth.ac.uk/38089/

Source: Web of Science (Lite)

Fractional Derivative Based Weighted Skip Connections for Satellite Image Road Segmentation

Authors: Pandey, H., Aroraa, S., Suman, H.K., Mathur, T. and Tiwari, K.

Journal: Neural Networks

Publisher: Elsevier

ISSN: 0893-6080

Abstract:

Segmentation of a road portion from a satellite image is challenging due to its complex background, occlusion, shadows, clouds, and other optical artifacts. One must combine both local and global cues for an accurate and continuous/connected road network extraction. This paper proposes a model using fractional derivative-based weighted skip connections on a densely connected convolutional neural network for road segmentation. Weights corresponding to the skip connections are determined using Grunwald-Letnikov fractional derivative. Fractional derivatives being non-local in nature incorporates memory into the system and thereby combine both local and global features. Experiments have been performed on two open source widely used benchmark databases viz. Massachusetts Road database (MRD) and Ottawa Road database (ORD). Both these datasets represent different road topography and network structure including varying road widths and complexities. Result reveals that the proposed system demonstrated better performance than the other state-of-the-art methods by achieving an F1-score of 0.748 and the mIoU of 0.787 at fractional order 0.4 on the MRD and a mIoU of 0.9062 at fractional order 0.5 on the ORD.

https://eprints.bournemouth.ac.uk/38089/

https://www.sciencedirect.com/journal/neural-networks

Source: Manual

Fractional derivative based weighted skip connections for satellite image road segmentation.

Authors: Arora, S., Suman, H.K., Mathur, T., Pandey, H.M. and Tiwari, K.

Journal: Neural networks : the official journal of the International Neural Network Society

Volume: 161

Pages: 142-153

eISSN: 1879-2782

ISSN: 0893-6080

DOI: 10.1016/j.neunet.2023.01.031

Abstract:

Segmentation of a road portion from a satellite image is challenging due to its complex background, occlusion, shadows, clouds, and other optical artifacts. One must combine both local and global cues for an accurate and continuous/connected road network extraction. This paper proposes a model using fractional derivative-based weighted skip connections on a densely connected convolutional neural network for road segmentation. Weights corresponding to the skip connections are determined using Grunwald-Letnikov fractional derivative. Fractional derivatives being non-local in nature incorporates memory into the system and thereby combine both local and global features. Experiments have been performed on two open source widely used benchmark databases viz. Massachusetts Road database (MRD) and Ottawa Road database (ORD). Both these datasets represent different road topography and network structure including varying road widths and complexities. Result reveals that the proposed system demonstrated better performance than the other state-of-the-art methods by achieving an F1-score of 0.748 and the mIoU of 0.787 at fractional order 0.4 on the MRD and a mIoU of 0.9062 at fractional order 0.5 on the ORD.

https://eprints.bournemouth.ac.uk/38089/

Source: Europe PubMed Central

Fractional Derivative Based Weighted Skip Connections for Satellite Image Road Segmentation

Authors: Aroraa, S., Suman, H.K., Mathur, T., Pandey, H. and Tiwari, K.

Journal: Neural Networks

Volume: 161

Pages: 142-153

Publisher: Elsevier

ISSN: 0893-6080

Abstract:

Segmentation of a road portion from a satellite image is challenging due to its complex background, occlusion, shadows, clouds, and other optical artifacts. One must combine both local and global cues for an accurate and continuous/connected road network extraction. This paper proposes a model using fractional derivative-based weighted skip connections on a densely connected convolutional neural network for road segmentation. Weights corresponding to the skip connections are determined using Grunwald-Letnikov fractional derivative. Fractional derivatives being non-local in nature incorporates memory into the system and thereby combine both local and global features. Experiments have been performed on two open source widely used benchmark databases viz. Massachusetts Road database (MRD) and Ottawa Road database (ORD). Both these datasets represent different road topography and network structure including varying road widths and complexities. Result reveals that the proposed system demonstrated better performance than the other state-of-the-art methods by achieving an F1-score of 0.748 and the mIoU of 0.787 at fractional order 0.4 on the MRD and a mIoU of 0.9062 at fractional order 0.5 on the ORD.

https://eprints.bournemouth.ac.uk/38089/

https://www.sciencedirect.com/journal/neural-networks

Source: BURO EPrints