Brain Slices Microscopic Detection Using Simplified SSD with Cycle-GAN Data Augmentation
Orderly automatic collection of brain slices on the silicon substrate is critical for understanding the working principle of the whole-brain neural network. Accurate and real-time brain slices detection with microscopic CCD is crucial for automatic collection of brain slices. To solve this task, an efficient simplified SSD detection model with Cycle-GAN data augmentation is presented in this paper. The proposed simplified SSD streamlines the detection network of the original SSD architecture, leading to a more rapid detection. Moreover, the proposed Cycle-GAN data augmentation method overcomes the limitation of training images. To verify the effectiveness of the proposed method, experiments are conducted with a self-made brain slices dataset. The experiment results suggest that, the proposed method has a good performance of rapidly detecting brain slices with only a small training dataset.
KeywordsMicroscopic object detection Deep learning Data augmentation
This work was supported in part by the National Natural Science Foundation of China under Grants 61873268, 61633016, in part by the Research Fund for Young Top-Notch Talent of National Ten Thousand Talent Program, in part by the Beijing Municipal Natural Science Foundation under Grant 4162066.
- 4.Perez, L., Wang, J.: The effectiveness of data augmentation in image classification using deep learning. arXiv preprint arXiv:1712.04621 (2017)
- 5.Goodfellow, I., et al.: Generative adversarial nets. In: Advances in Neural Information Processing Systems, pp. 2672–2680 (2014)Google Scholar
- 6.Wei, L., Zhang, S., Gao, W., Tian, Q.: Person transfer GAN to bridge domain gap for person re-identification. arXiv preprint arXiv:1711.08565 (2017)
- 7.Mariani, G., Scheidegger, F., Istrate, R., Bekas, C., Malossi, C.: BAGAN: data augmentation with balancing GAN. arXiv preprint arXiv:1803.09655 (2018)
- 8.Mao, Y., Yin, Z., Schober, J.M.: Iteratively training classifiers for circulating tumor cell detection. In: IEEE 12th International Symposium on Biomedical Imaging, pp. 190–194 (2015)Google Scholar
- 9.Holmström, O., et al.: Point-of-care mobile digital microscopy and deep learning for the detection of soil-transmitted Helminths and Schistosoma haematobium. Glob. Health. Action. 10, 1337325 (2017)Google Scholar
- 10.Yang, S., Fang, B., Tang, W., Wu, X., Qian, J., Yang, W.: Faster R-CNN based microscopic cell detection. In: International Conference on Security, Pattern Analysis, and Cybernetics, pp. 345–350 (2017)Google Scholar
- 11.Hung, J., Carpenter, A.: Applying faster R-CNN for object detection on malaria images. In: IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 808–813 (2017)Google Scholar
- 12.Zhang, J., Hu, H., Chen, S., Huang, Y., Guan, Q.: Cancer cells detection in phase-contrast microscopy images based on Faster R-CNN. In: 9th International Symposium on Computational Intelligence and Design, pp. 363–367 (2016)Google Scholar
- 14.Yi, J., Wu, P., Hoeppner, D.J., Metaxas, D.: Fast neural cell detection using light-weight SSD neural network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 108–112 (2017)Google Scholar
- 15.Zhu, J.-Y., Park, T., Isola, P., Efros, A.A.: Unpaired image-to-image translation using cycle-consistent adversarial networks. arXiv preprint arXiv:1703.10593 (2017)