Abstract
Since variation of car’s types, direction classification of crash test photos for multiple type of vehicles is a big challenge to office automation. Moreover, because of the similarity of images before and after crash test, semantic identification of these two classes is also difficult. Inspired by recent advances in large kernel CNNs, in this paper, we introduce a 31 × 31 extra-large convolutional kernel to gain more effective receptive field and makes the model more powerful for semantic segmentation. To meet the requirement to generate testing report, totally 14 classes were applied in our experiment. The accuracy of 97.1% was obtained on a self-build photo dataset.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Ioffe, S., Szegedy, C.: Batch normalization: Accelerating deep network training by reducing internal covariate shift. In: International Conference on Machine Learning, pp. 448–456 (2015)
Veit, A., Wilber, M.J., Belongie, S.: Residual networks behave like ensembles of relatively shallow networks. In: Advances in Neural Information Processing Systems, pp. 550–558 (2016)
Dong, Y., Cordonnier, J.-B., Loukas, A.: Attention is not all you need: Pure attention loses rank doubly exponentially with depth. arXiv preprint arXiv:2103.03404, (2021)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)
Pasupa, K., Kittiworapanya, P., Hongngern, N., et al.: Evaluation of deep learning algorithms for semantic segmentation of car parts. Complex Intell. Syst. 8, 3613–3625 (2022)
Hendrycks, D., Gimpel, K.: Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415, (2016)
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., Chen, L.-C.: Mobilenetv2: inverted residuals and linear bottlenecks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4510–4520 (2018)
Ding, X., Zhang, X., Ma, N., Han, J., Ding, G., Sun, J.: Repvgg: making vggstyle convnets great again. In :Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 13733–13742 (2021)
Doll ́ar, P., Singh, M., Girshick, R.: Fast and accurate model scaling. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 924–932 (2021)
Radosavovic, I., Kosaraju, R.P., Girshick, R., He, K., Doll ́ar, P.: Designing network design spaces. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10428–10436 (2020)
Dong, Y., Cordonnier, J.-P., Loukas, A.: Attention is not all you need: Pure attention loses rank doubly exponentially with depth. arXiv preprint arXiv:2103.03404, (2021)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, J., Xiao, Q., Liu, J., Huang, Z., Wang, T., Li, G. (2023). A Novel Convolutional Neural Network with Large Kernel for Classification of Crash Test Photos. In: Lu, H., Blumenstein, M., Cho, SB., Liu, CL., Yagi, Y., Kamiya, T. (eds) Pattern Recognition. ACPR 2023. Lecture Notes in Computer Science, vol 14406. Springer, Cham. https://doi.org/10.1007/978-3-031-47634-1_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-47634-1_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-47633-4
Online ISBN: 978-3-031-47634-1
eBook Packages: Computer ScienceComputer Science (R0)