Binary Classification of Images for Applications in Intelligent 3D Scanning
Three-dimensional (3D) scanning techniques based on photogrammetry, also known as Structure-from-Motion (SfM), require many two-dimensional (2D) images of an object, obtained from different viewpoints, in order to create its 3D reconstruction. When these images are acquired using closed-space 3D scanning rigs, which are composed of large number of cameras fitted on multiple pods, flash photography is required and image acquisition must be well synchronized to avoid the problem of ‘misfired’ cameras. This paper presents an approach to binary classification (as ‘good’ or ‘misfired’) of images obtained during the 3D scanning process, using four machine learning methods—support vector machines, artificial neural networks, k-nearest neighbors algorithm, and random forests. Input to the algorithms are histograms of regions determined to be of interest in the detection of image misfires. The considered algorithms are evaluated based on the prediction accuracy that they achieved on our dataset. The average prediction accuracy of 94.19% is obtained using the random forests approach under cross-validation. Therefore, the application of the proposed approach allows the development of an ‘intelligent’ 3D scanning system which can automatically detect camera misfiring and repeat the scanning process without the need for human intervention.
KeywordsMachine learning Image processing Image classification Binary classification Decision trees Random forest 3D scanning Photogrammetry Structure-from-motion
The authors would like to thank Doob Group AG for the support and dataset provided for this research. The reported research is also partly supported by the Ministry of Education, Science, and Technological Development of the Republic of Serbia, projects TR32044 (2011–2017), ON174026 (2011–2017), and III44006 (2011–2017).
- 1.Hartley, R., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press (2004)Google Scholar
- 2.Luhmann, T., Robson, S., Kyle, S., Boehm, J.: Close-Range Photogrammetry and 3D Imaging, revised 1st edn. De Gruyter (2013)Google Scholar
- 3.Theodoridis, S., Koutroumbas, K.: Pattern Recognition, 4th edn. Elsevier (2008)Google Scholar
- 4.Menze, B.H., Kelm, B.M., Masuch, R., Himmelreich, U., Bachert, P., Petrich, W., Hamprecht, F.A.: A comparison of random forest and its Gini importance with standard chemometric methods for the feature selection and classification of spectral data. BMC Bioinformatics 10:213 (2009). doi: 10.1186/1471-2105-10-213
- 5.Random forest. https://en.wikipedia.org/wiki/Random_forest. Last accessed 13 Apr 2017
- 6.Basten, K.: Classifying Landsat Terrain Images via Random Forests. Radboud University, Nijmegen, Netherlands (2016)Google Scholar
- 8.Melgani, F., Bruzzone, L.: Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. Geosci. Remote Sens. 42(8), 1778–1790 (2004)Google Scholar
- 10.Boiman, O., Shechtman, E., Irani, M.: In defense of Nearest-Neighbor based image classification. In: Proceedings IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Anchorage, USA, pp. 1–8 (2008)Google Scholar
- 11.OpenCV library, www.opencv.org. Last accessed 13 Apr 2017
- 12.Scikit-learn, http://scikit-learn.org/stable/. Last accessed 26 Apr 2017
- 13.Keras, https://keras.io/. Last accessed 26 Apr 2017