Recognition of Assembly Parts by Convolutional Neural Networks
The paper describes the experiments with the use of deep neural networks (CNN) for robust identification of assembly parts (screws, nuts) and assembly features (holes), to speed up any assembly process with augmented reality application. The simple image processing tasks with static camera and recognized parts can be handled by standard image processing algorithms (threshold, Hough line/circle detection and contour detection), but the augmented reality devices require dynamic recognition of features detected in various distances and angles. The problem can be solved by deep learning CNN which is robust to orientation, scale and in cases when element is not fully visible. We tested two pretrained CNN models Mobilenet V1 and SSD Fast RCNN Inception V2 SSD extension have been tested to detect exact position. The results obtained were very promising in comparison to standard image processing techniques.
KeywordsDeep learning Object recognition Augmented reality
This work was supported by the Agency for Research and Development under the contract no. APVV-15-0602.
- 2.Takaseki, A.R., Nagashima, R., Kashima, H., Okazaki, T.: Development of anchoring support system using with AR toolkit. In: 2015 7th International Conference on Emerging Trends in Engineering and Technology, pp. 123–127 (2015)Google Scholar
- 4.Chuck, C., Laskey, M., Krishnan, S., Joshi, R., Fox, R., Goldberg, K., et al:. Statistical data cleaning for deep learning of automation tasks from demonstrations. In: 13th IEEE Conference on Automation Science and Engineering (IEEE CASE), 20–23 August, pp. 1142–1149. IEEE, Xian (2017)Google Scholar
- 8.Patil, K., Kulkarni, M., Sriraman, A., Karande, S.: Deep learning based car damage classification. In: 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 50–54 (2017)Google Scholar
- 9.Liang, J., Mahler, J., Laskey, M., Li, P.S., Goldberg, K.: Using dVRK teleoperation to facilitate deep learning of automation tasks for an industrial robot. In: 13th IEEE Conference on Automation Science and Engineering (IEEE CASE), 20–23 August, pp. 1–8. IEEE, Xian (2017)Google Scholar
- 12.Manzanera, A., Nguyen, T.P., Xu, X.L.: Line and circle detection using dense one-to-one Hough transforms on greyscale images. EURASIP J. Image Video Process. 18 (2016)Google Scholar