Gaussian Compound Feature Descriptor and Contourlet Transform for Image Registration in 3D Reconstruction

Huan Liu; Genfu Xiao; Wei Peng

doi:10.6180/jase.201809_21(3).0013

Gaussian Compound Feature Descriptor and Contourlet Transform for Image Registration in 3D Reconstruction

Huan Liu¹, Genfu Xiao This email address is being protected from spambots. You need JavaScript enabled to view it.² and Wei Peng³

¹College of Electronic and Information Engineering, Jinggangshan University & Key Laboratory of Watershed Ecology and Geographical Environment Monitoring NASG, Jinggangshan University, Ji’an, Jiangxi 343009, P.R. China
²College of Mechanical and Electronic, Jianggangshan University, Ji’an, Jiangxi 343009, P.R. China
³College of Architectural and Civil Engineering, Jianggangshan University, Ji’an, Jiangxi 343009, P.R. China

Received: April 11, 2017
Accepted: April 27, 2018
Publication Date: September 1, 2018

Download Citation: ||https://doi.org/10.6180/jase.201809_21(3).0013

ABSTRACT

This paper presents a new method for multi-view images registration based on stereo vision system. Our aim is to recover the surface shape of the subject accurately and efficiently in spite of the influences derived from illumination variation, blur affection and image transformation on the 2D images. For this purpose, we devote to developing an innovative stereo registration algorithm. In the first phase, a novel feature descriptor is constructed by adding multi-scale Gaussian parameters into the illumination-robust and anti-blur combined moment invariants in fusion of the pixel gray and gradient. The new Gaussian combined moment invariants are calculated on the multi-scale low frequency sub-band by Contourlet transform. Meanwhile, grid entropy was computed on the multidirection high frequency sub-band as to get the structure characteristics of the image. Then a novel compound feature descriptor was presented by a combination of the Gaussian moment invariants and grid entropy. It is applied to conduct the similarity measure for the initial image registration. In the second phase, the bidirectional matching strategy with strict geometric constraints composed of the distance and the slope between matching pairs is proposed for eliminating the incorrect matching pairs in the initial image registration. Consequently, the correct matching pairs are obtained at this stage.The experimental results reveal that both the accuracy and the efficiency of our approach are superior to those of SIFT and SURF. Finally, 3D cloud data and 3D model of the subject are achieved.

Keywords: Multi-scale and Multi-direction, Gaussian Combined Moment Invariant, Grid Entropy, Feature Registration, Bidirectional Geometric Constraint, 3D Reconstruction

REFERENCES

[1] Wang, X., Feng, B., Bai, X., et al., “Bag of Contourlet Fragments for Robust Shape Classification,” Pattern Recognition, Vol. 47, No. 6, pp. 21162125 (2014). doi: 10.1016/j.patcog.2013.12.008
[2] Yao, C., Bai, X. and Liu, W., “A Unified Framework for Multi-oriented Text Detection and Recognition,” IEEE Transactions on Image Processing, Vol. 23, No. 11, pp. 47374749 (2014). doi: 10.1109/TIP.2014.2353813
[3] Gong, M., Zhao, J., Liu, Q., et al., “Change Detection in Synthetic Aperture Radar Images Based on Deep Neural Networks,” IEEE Transactions on Neural Networks and Learning Systems, Vol. 27, No. 1, pp. 125 138 (2015). doi: 10.1109/TNNLS.2015.2435783
[4] Guo, X. and Cao, X., “Good Match Exploration Using Triangle Constraint,” Pattern Recognition Letters, Vol. 33, No. 7, pp. 872881 (2012). doi: 10.1016/j.patrec.2011.08.021
[5] Ma, J., Zhao, J., Tian, J., et al., “Robust Estimation of Nonrigid Transformation for Point Set Registration,” Computer Vision and Pattern Recognition, Vol. 9, No. 4, pp. 21472154 (2013). doi: 10.1109/CVPR.2013.279
[6] Tharkar, K., Kapadia, D. and Natali, F., “Implementation and Analysis of TemplateMatching for Image Registration on Devkit-8500D,” Optik-International Journal for Light and Electron Optics, Vol. 130, No. 2, pp. 935944 (2017). doi: 10.1016/j.ijleo.2016.11.057
[7] Liu, H., Hao, K. R. and Ding, Y. S., “New Anti-blur and Illumination-robust Combined Invariant for Stereo Vision in Human Belly Reconstruction,” Imaging Science Journal, Vol. 62, No. 5, pp. 251264 (2014).doi: 10.1179/1743131X13Y.0000000061
[8] Xu, X. C., Zheng, Z. Z., Xu, A. G., et al., “An Optimized Method for Terrain Reconstruction Based on Descent Images,” Journal of Engineering and Technological Sciences, Vol. 48, No. 1, pp. 3148 (2016). doi:10.5614/j.eng.technol.sci.2016.48.1.4
[9] Liang, Zh., Jinfa, X. and Qingyuan, X., “Pose Estimation Algorithm and Verification Based on Binocular Stereo Vision for Unmanned Aerial Vehicle,” Journal of Harbin Institute of Technology, Vol. 46, No. 5, pp. 6672 (2014).
[10] Xia, X. H., Dang, G. and Yao, Y. S., “Image Registration Model and Algorithm for Multi-focus Images,” Pattern Recognition Letters, Vol. 86, No. 15, pp. 26 30 (2017). doi: 10.1016/j.patrec.2016.12.005
[11] Wu, Y., Ma, W., Gong, M., et al., “ANovel Point-matching Algorithm Based on Fast Sample Consensus for Image Registration,” IEEE Geoscience and Remote Sensing Letters, Vol. 12, No. 1, pp. 4347 (2015). doi:10.1109/LGRS.2014.2325970
[12] Zitoví, B. and Flusser, J., “Image Registration Methods: a Survey,” Image and Vision Computing, Vol. 21, No. 11, pp. 9771000 (2003). doi: 10.1016/S0262-8856(03)00137-9
[13] Lowe, D. G., “Distinctive Image Features from Scale invariant Keypoints,” International Journal of Computer Vision, Vol. 60, No. 2, pp. 91110 (2004). doi:10.1023/B:VISI.0000029664.99615.94
[14] Zhang, K., Li, X. and Zhang, J., “A Robust Point matching Algorithm for Remote Sensing Image Registration,” IEEE Geoscience and Remote Sensing Letters, Vol. 11, No. 2, pp. 469473 (2014). doi: 10.1109/LGRS.2013.2267771
[15] Li, B. and Ye, H., “RSCJ: Robust Sample Consensus Judging Algorithm for Remote Sensing Image Registration,” IEEE Geoscience and Remote Sensing Letters, Vol. 9, No. 4, pp. 574578 (2012). doi:10.1109/LGRS.2011.2175434
[16] Li, Q., Qi, S. and Shen, Y., “Multispectral Image Alignment with Nonlinear Scale-invariant Keypoint and Enhanced Local Feature Matrix,” IEEE Geoscience and Remote Sensing Letters, Vol. 12, No. 7, pp. 15 (2015). doi: 10.1109/LGRS.2015.2412955
[17] Lv, G. H., Teng, S. W. and Lu, G. J., “Enhancing SIFT based Image Registration Performance by Building and Selecting Highly Discriminating Descriptors,” Pattern Recognition Letters, Vol. 84, No. 1, pp. 156162 (2016). doi: 10.1016/j.patrec.2016.09.011
[18] Huang, L. Q., Chen, C. G. and Shen, H. H., “Adaptive Registration Algorithm of Color Image Based on SURF,” Measurement, Vol. 66, No. 4, pp. 118124 (2015). doi:10.1016/j.measurement.2015.01.011
[19] Patel, M. I., Thakar, V. K. and Shah, S. K., “Image Registration of Satellite Images with Varying Illumination Level Using HOG Descriptor Based SURF,” Procedia Computer Science, Vol. 93, pp. 382388 (2016). doi: 10.1016/j.procs.2016.07.224
[20] Dellinger, J. F., Delon, J., Gousseau, Y., et al., “Sarsift: a Sift-like Algorithm for Sar Iamges,” IEEE Transactions on Geoscience and Remote Sensing, Vol. 53, No. 1, pp. 453466 (2015). doi: 10.1109/TGRS.2014.2323552
[21] Wu, Y., Gong, M. G., Jia, J., et al., “Reomte Sensing Image Registration with Spatial Restraint Based on Moment Invariants and Fast Generalized Fuzzy Clustering,” 2015 Conference on Technologies and Applications of Artificial Intelligenc, Nov. 2022, pp. 97 104 (2015). doi: 10.1109/TAAI.2015.7407062
[22] Liu, Y., He, F., Zhu, X., et al., “The Improved Characteristics of Bionic Gabor Representations by Combining with SIFT Key-points for Iris Recognition,” Journal of Bionic Engineering, Vol. 12, No. 3, pp. 504517(2015). doi: 10.1016/S1672-6529(14)60141-4
[23] Peng, Y. and Guo, W. Y., “Feature Extraction Using Dual-ftree Complex Wavelet Transform and Gray Level Co-occurrence Matrix,” Neurocomputing, Vol. 197, pp. 212 220 (2016). doi: 10.1016/j.neucom.2016.02.061
[24] Patil, H. Y., Kothari, A. G. and Bhurchandi, K. M., “Expression Invariant Face Recognition Using Local Binary Pattern and Contourlet Transform,” International Journal of Light and Electron Optics, Vol. 127, No. 5, pp. 26702678 (2016). doi: 10.1016/j.ijleo.2015.11.187
[25] Cai, J. J., Cheng, Q. M., Peng, M. J., et al., “Fusion of Infrared and Visible Images Based on Nonsubsampled Contourlet Transform and Sparse K-SVD Dictionary Learning,” Infrared Physics & Technology, Vol. 82, pp. 8595 (2017). doi: 10.1016/j.infrared.2017.01.026
[26] Srivastava, A., Bhateja, V. and Moin, A., “Combination of PCA and Contourlets for Multispectral iMage Fusion,” Proceedings of the International Conference on Data Engineering and Communication Technology, pp. 577585 (2017). doi: 10.1007/978-981-10-1678-3_55
[27] Liu, L., Jia, Z. and Kasabov, N., “A Remote Sensing Image Enhancement Method Using Mean Filter and Unsharp Masking in Non-subsampled Contourlet Transform Domain,” Transactions of the Institute of Measurement and Control, Vol. 39, No. 2, pp. 183193 (2017). doi: 10.1177/0142331215604210
[28] Duan, G. Y., Yang, J. and Yang, Y. L., “Content-based Image Retrieval Research,” Physics Procedia, Vol. 22, No. 11, pp. 471477 (2011). doi: 10.1016/j.phpro.2011.11.073
[29] Serief, C., Barkat, M., et al., “Robust Feature Points Extraction for Image Registration Based on the Non subsampled Contourlet Transform,” Interational Journal of Electronics and Communications, Vol. 63, pp. 148152 (2009). doi: 10.1016/j.aeue.2007.11.005
[30] Border, S. and Border, U., “Image Retrieval Using Contourlet Transform,” International Journal of Computer Application, Vol. 11, No. 34, pp. 3743 (2011).
[31] Dong, Y. S. and Ma,J. W., “FeatureExtraction through Contourlet Subband Clustering for Texture Classification,” Neurocomputing, Vol. 116, No. 10, pp. 157164 (2013). doi: 10.1016/j.neucom.2011.12.059
[32] Sobolewski, T., Messer, N., et al., “Contourlet Image Preprocessing for Enhanced Control Point Selection in Airborne Image Registration,” Applied Imagery Pattern Recognition Workshop, pp. 16 (2016).
[33] Ma, J., Zhao, J., Tian, A., et al., “Robust Point Matching via Vector Field Consensus,” IEEE Transactions on Image Processing, Vol. 23, No. 4, pp. 17061721 (2014). doi: 10.1109/TIP.2014.2307478
[34] Pang, S., Xue, J., Tian, Q., et al., “Exploiting Local Linear Geometric Structure for Identifying Correct Matches,” Computer Vision and Image Understanding, Vol. 128, No. 11, pp. 5164 (2014). doi: 10.1016/j.cviu.2014.06.006
[35] Ma, J., Ma, Y., Zhao, J., et al., “Image Feature Matching via Progressive Vector Field Consensus,” IEEE Signal Processing Letters, Vol. 22, No. 6, pp. 767771 (2015). doi: 10.1109/LSP.2014.2358625
[36] Fischler, M. A. and Bolles, R. C., “Random Sample Consensus: a Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography,” Communications of the ACM, Vol. 24, No. 6, pp. 381395 (1981). doi: 10.1145/358669.358692
[37] Do, M. N. and Vetterli, M. V., “The Contourlet Transform: a Efficient Directional Multiresolution Image Representation,” IEEE Transactions on Image Processing, Vol. 14, No. 12, pp. 20912106 (2005). doi:10.1109/TIP.2005.859376