Abstract
Deep CNN’s have achieved an excellent performance in computer vision and image processing methods, designating them as a state-of-art in this domain. CNN based applications have achieved tremendous advancement towards vision computing with high dimensional object labelling in images. The complex nature of High Dimensional (HD) images limits the performance of CNN’s. In high dimensional feature space, the pixel-based image labelling is a complex problem for the parsing of objects in an image. To overcome this issue, we have studied a two-stage end-to-end framework that uses manifold embedding based patch-wise CNN architecture to extract the features and classify the image for labelled classes. We have investigated the deep-features with an information fusion technique for low dimensional feature space compression by using pre-trained CNNs and spatiality preserving manifold embedding in the first stage. The cost of pixel-based labelling in HD feature space is very high, so researchers have tried to encapsulate maximum information within the minimum image size. Therefore, in this stage, we have first increased the valuable information by concatenating the deep spatial features and then embedding the massive dataset by using manifold preservation. In stage-2, the image patches are extracted and passed into three layers of convolution-pooling pair and two layers of fully connected pair using parameter tuning. The training dataset is prepared in the form of pixel-label pairs. Subsequently, the proposed method has been evaluated on publicly available images and compared with the previously proposed schemes. The proposed method has outperformed the previous techniques in accuracy and computation time with a significant margin.
Similar content being viewed by others
References
Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, Devin M, Ghemawat S, Irving G, Isard M, Kudlur M, Levenberg J, Monga R, Moore S, Murray DG, Steiner B, Tucker P, Vasudevan V, Warden P, Wicke M, Yu Y, Zheng X (2016) Tensorflow: A system for large-scale machine learning. In: Proceedings of the 12th USENIX conference on operating systems design and implementation, USENIX association, Berkeley, CA, USA, OSDI’16, pp 265–283, http://dl.acm.org/citation.cfm?id=3026877.3026899
Li L, Ge H, Gao J, Zhang Y, Tong Y, Sun J (2019) A novel geometric mean feature space discriminant analysis method for hyperspectral image feature extraction. Neural Process Lett 51(1):515–542. https://doi.org/10.1007/s11063-019-10101-0
Bhandari AK, Kumar A, Singh GK, Soni V (2016) Performance study of evolutionary algorithm for different wavelet filters for satellite image denoising using sub-band adaptive threshold. J Exp Theor Artif Intell 28(1–2):71–95. https://doi.org/10.1080/0952813X.2015.1020518
Bian X, Zhang T, Zhang X, Yan L, Li B (2013) Clustering-based extraction of near border data samples for remote sensing image classification. Cogn Comput 5(1):19–31. https://doi.org/10.1007/s12559-012-9147-2
Cai D, He X, Han J (2007) Spectral regression: a unified subspace learning framework for content-based image retrieval. In: ACM multimedia
Cai Z, Shao L (2018) Rgb-d scene classification via multi-modal feature learning. Cogn Comput. https://doi.org/10.1007/s12559-018-9580-y
Cao X, Zhou F, Xu F, Meng D, Xu Z, Paisley J (2018) Hyperspectral image classification with Markov random fields and a convolutional neural network. IEEE Trans Image Process 27(5):2354–2367. https://doi.org/10.1109/TIP.2018.2799324
Cheng G, Han J, Lu X (2017) Remote sensing image scene classification: Benchmark and state of the art. Proceedings of the IEEE 105(10):1865–1883. https://doi.org/10.1109/JPROC.2017.2675998
Cheng G, Li Z, Han J, Yao X, Guo L (2018) Exploring hierarchical convolutional features for hyperspectral image classification. IEEE Trans Geosci Remote Sens 56(11):6712–6722. https://doi.org/10.1109/TGRS.2018.2841823
Ergul U, Bilgin G (2019) Hckboost: Hybridized composite kernel boosting with extreme learning machines for hyperspectral image classification. Neurocomputing 334:100–113. https://doi.org/10.1016/j.neucom.2019.01.010
Han D (2013/03) Comparison of commonly used image interpolation methods. In: Proceedings of the 2nd international conference on computer science and electronics engineering. Atlantis Press, https://doi.org/10.2991/iccsee.2013.391
Hu W, Huang Y, Wei L, Zhang F, Li H (2015) Deep convolutional neural networks for hyperspectral image classification. J Sens 258619:12. https://doi.org/10.1155/9161
Imani M (2018) Anomaly detection using morphology-based collaborative representation in hyperspectral imagery. Eur J Remote Sens 51(1):457–471. https://doi.org/10.1080/22797254.2018.1446727
Knöbelreiter P, Reinbacher C, Shekhovtsov A, Pock T (2017) End-to-end training of hybrid CNN-CRF models for stereo. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 1456–1465. https://doi.org/10.1109/CVPR.2017.159
Le THN, Duong CN, Han L, Luu K, Quach KG, Savvides M (2018) Deep contextual recurrent residual networks for scene labeling. Pattern Recognit 80:32–41. https://doi.org/10.1016/j.patcog.2018.01.005
Li D, Tian Y (2018) Survey and experimental study on metric learning methods. Neural Netw 105:447–462. https://doi.org/10.1016/j.neunet.2018.06.003
Li T, Leng J, Kong L, Guo S, Bai G, Wang K (2019) Dcnr: deep cube cnn with random forest for hyperspectral image classification. Multimed Tools Appl 78(3):3411–3433. https://doi.org/10.1007/s11042-018-5986-5
Li W, Du Q, Zhang B (2015) Combined sparse and collaborative representation for hyperspectral target detection. Pattern Recognit 48(12):3904–3916. https://doi.org/10.1016/j.patcog.2015.05.024
Li W, Wu G, Zhang F, Du Q (2017) Hyperspectral image classification using deep pixel-pair features. IEEE Trans Geosci Remote Sens 55(2):844–853. https://doi.org/10.1109/TGRS.2016.2616355
Li W, Ding W, Sadasivam R, Cui X, Chen P (2019b) His-gan: a histogram-based gan model to improve data generation quality. Neural Netw 119:31–45. https://doi.org/10.1016/j.neunet.2019.07.001
Li Y, Lu BL (2009) Feature selection based on loss-margin of nearest neighbour classification. Pattern Recognit 42(9):1914–1921. https://doi.org/10.1016/j.patcog.2008.10.011
Gao H, Lin S, Li C, Yang Y (2018) Application of hyperspectral image classification based on overlap pooling. Neural Process Lett 49(3):1335–1354. https://doi.org/10.1007/s11063-018-9876-7
Ma J, Yuan Y (2019) Dimension reduction of image deep feature using pca. J Vis Commun Image Represent 63:102578. https://doi.org/10.1016/j.jvcir.2019.102578
Ma X, Liu W, Tao D, Zhou Y (2019) Ensemble p-Laplacian regularization for scene image recognition. Cogn Comput. https://doi.org/10.1007/s12559-019-09637-z
Marinoni A, Gamba P (2017) Unsupervised data driven feature extraction by means of mutual information maximization. IEEE Trans Comput Imaging 3(2):243–253. https://doi.org/10.1109/TCI.2017.2669731
Menassel R, Nini B, Mekhaznia T (2018) An improved fractal image compression using wolf pack algorithm. J Exp Theor Artif Intell 30(3):429–439. https://doi.org/10.1080/0952813X.2017.1409281
Li L, Ge H, Gao J, Zhang Y (2018) Hyperspectral image feature extraction using maclaurin series function curve fitting. Neural Process Lett 49(1):357–374. https://doi.org/10.1007/s11063-018-9825-5
Nasrabadi NM (2014) Hyperspectral target detection: an overview of current and future challenges. IEEE Signal Process Mag 31(1):34–44. https://doi.org/10.1109/MSP.2013.2278992
Cahill ND, Chew SE, Wenger PS (2015) Spatial-spectral dimensionality reduction of hyperspectral imagery with partial knowledge of class labels. https://doi.org/10.1117/12.2177139
Pan E, Mei X, Wang Q, Ma Y, Ma J (2020) Spectral-spatial classification for hyperspectral image based on a single gru. Neurocomputing. https://doi.org/10.1016/j.neucom.2020.01.029
Peng J, Jiang X, Chen N, Fu H (2019) Local adaptive joint sparse representation for hyperspectral image classification. Neurocomputing 334:239–248. https://doi.org/10.1016/j.neucom.2019.01.034
Priego B, Duro RJ, Chanussot J (2017) 4dcaf: a temporal approach for denoising hyperspectral image sequences. Pattern Recognit 72:433–445. https://doi.org/10.1016/j.patcog.2017.07.023
Rahimi SA, Sajedi H (2019) Monitoring air pollution by deep features and extreme learning machine. J Exp Theor Artif Intell 31(4):517–531. https://doi.org/10.1080/0952813X.2019.1572658
Sharma A, Liu X, Yang X (2018) Land cover classification from multi-temporal, multi-spectral remotely sensed imagery using patch-based recurrent neural networks. Neural Netw 105:346–355. https://doi.org/10.1016/j.neunet.2018.05.019
Srivastava V, Biswas B (2019) Cnn-based salient features in hsi image semantic target prediction. Connect Sci. https://doi.org/10.1080/09540091.2019.1650330
Srivastava V, Biswas B (2019) Deep cnn feature fusion with manifold learning and regression for pixel classification in hsi images. J Exp Theor Artif Intell. https://doi.org/10.1080/0952813X.2019.1647566
Srivastava V, Biswas B (2019) A subspace regression and two phase label optimization for high dimensional image classification. Multimed Tools Appl. https://doi.org/10.1007/s11042-019-08477-1
Ho Tin Kam (1998) The random subspace method for constructing decision forests. IEEE Trans Pattern Anal Mach Intell 20(8):832–844. https://doi.org/10.1109/34.709601
Trentin E, Cattoni R (1999) Learning perception for indoor robot navigation with a hybrid hidden Markov model/recurrent neural networks approach. Connect Sci 11(3–4):243–265. https://doi.org/10.1080/095400999116241
Tu B, Li N, Fang L, Fei H, He D (2018) Classification of hyperspectral images via weighted spatial correlation representation. J Vis Commun Image Represent 56:160–166. https://doi.org/10.1016/j.jvcir.2018.09.010
Xie W, Li Y, Hu J, Chen DY (2018) Trainable spectral difference learning with spatial starting for hyperspectral image denoising. Neural Netw 108:272–286. https://doi.org/10.1016/j.neunet.2018.08.021
Yao Y, Guo P, Xin X, Jiang Z (2014) Image fusion by hierarchical joint sparse representation. Cogn Comput 6(3):281–292. https://doi.org/10.1007/s12559-013-9235-y
Zhang A, Liu S, Sun G, Huang H, Ma P, Rong J, Ma H, Lin C, Wang Z (2018) Clustering of remote sensing imagery using a social recognition-based multi-objective gravitational search algorithm. Cogn Comput. https://doi.org/10.1007/s12559-018-9582-9
Zhang L, Barnden J (2012) Affect sensing using linguistic, semantic and cognitive cues in multi-threaded improvisational dialogue. Cogn Comput 4(4):436–459. https://doi.org/10.1007/s12559-012-9170-3
Zhang P, He H, Gao L (2019) A nonlinear and explicit framework of supervised manifold-feature extraction for hyperspectral image classification. Neurocomputing 337:315–324. https://doi.org/10.1016/j.neucom.2019.01.077
Zhe X, Chen S, Yan H (2019) Directional statistics-based deep metric learning for image classification and retrieval. Pattern Recognit 93:113–123. https://doi.org/10.1016/j.patcog.2019.04.005
Zhu X, Zhang X, Zhang XY, Xue Z, Wang L (2019) A novel framework for semantic segmentation with generative adversarial network. J Vis Commun Image Represent 58:532–543. https://doi.org/10.1016/j.jvcir.2018.11.020
Zhu X, Zuo J, Ren H (2020) A modified deep neural network enables identification of foliage under complex background. Connect Sci 32(1):1–15. https://doi.org/10.1080/09540091.2019.1609420
Peng Y, Long X, Lu BL (2014) Graph based semi-supervised learning via structure preserving low-rank representation. Neural Process Lett 41(3):389–406. https://doi.org/10.1007/s11063-014-9396-z
Tzelepi M, Tefas A (2019) Class-specific discriminant regularization in real-time deep CNN models for binary classification problems. Neural Process Lett 51(2):1989–2005. https://doi.org/10.1007/s11063-019-10156-z
Venugopal N (2020) Automatic semantic segmentation with DeepLab dilated learning network for change detection in remote sensing images. Neural Process Lett. https://doi.org/10.1007/s11063-019-10174-x
Zhang Z, Zhang Y, Liu G, Tang J, Yan S, Wang M (2020) Joint label prediction based semi-supervised adaptive concept factorization for robust data representation. IEEE Trans Knowl Data Eng 32(5):951–970. https://doi.org/10.1109/TKDE.2019.2893956
Zhang Y, Zhang Z, Li S, Qin J, Liu G, Wang M, Yan S (2019) Unsupervised nonnegative adaptive feature extraction for data representation. IEEE Trans Knowl Data Eng 31(12):2423–2440. https://doi.org/10.1109/TKDE.2018.2877746
Zhang Z, Li F, Zhao M, Zhang L, Yan S (2017) Robust neighborhood preserving projection by nuclear/L2,1-norm regularization for image feature extraction. IEEE Trans Image Process 26(4):1607–1622. https://doi.org/10.1109/TIP.2017.2654163
Acknowledgements
We are thankful to Prof. P.Gamba from Pavia University, Italy, for providing the ROSIS dataset and Prof. Landgrebe of Purdue University for contributing to the Indian Pines data set. The Salinas Valley dataset has obtained from “Grupo de Inteligencia Computacional (GIC) Buscar”. We are thankful to Dr. Wei Li from the School of Information and Electronics, Beijing Institute of Technology, for providing the code for CNN-PPF methods in comparative analysis. We would like to thank Cao et al, Department of Mathematics and Statistics,Xi’an Jiaoting University, China, for providing the python function codes to design the customised CNN architecture.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Srivastava, V., Biswas, B. Manifold Preserving CNN for Pixel-Based Object Labelling in Images for High Dimensional Feature spaces. Neural Process Lett 53, 607–635 (2021). https://doi.org/10.1007/s11063-020-10415-4
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-020-10415-4