Abstract
Many industrial object-sorting applications leverage benefits of hyperspectral imaging technology. Design of object sorting algorithms is a challenging pattern recognition problem due to its multi-level nature. Objects represented by sets of pixels/spectra in hyperspectral images are to be allocated into pre-specified sorting categories. Sorting categories are often defined in terms of lower-level concepts such as material or defect types. This paper illustrates the design of two-stage sorting algorithms, learning to discriminate individual pixels/spectra and fusing the per-pixel decisions into a single per-object outcome. The paper provides a case-study on algorithm design in a real-world industrial sorting problem. Four groups of algorithms are studied varying the level of prior knowledge about the sorting problem. Apart of the sorting accuracy, the algorithm execution speed is estimated assuming an ideal implementation. Relating these two performance criteria allows us to discuss the accuracy/speed trade-off of different algorithms.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
We adopt the sigmoidal scaling of classifier outputs used in PRTools toolbox [2]. The sigmoid bias is fixed to zero and the slope parameter is estimated on the training set using the maximum-likelihood estimator.
References
Cheng, X.: Hyperspectral imaging and pattern recognition technologies for real time fruit safety and quality inspection. Ph.D. thesis, University of Maryland (2004)
Duin, R.P.W., Juszczak, P., de Ridder, D., Paclík, P., Pekalska, E., Tax, D.M.J.: PR-Tools 4.0, a Matlab toolbox for pattern recognition. Tech. rep., ICT Group, TU Delft, The Netherlands (2004) http://www.prtools.org
Duin, R.P.W., de Ridder, D., Tax, D.M.J.: Experiments with object based discriminant functions; a featureless approach to pattern recognition. Pattern Recognit. Lett. 18(11–13), 1159–1166 (1997)
Kong, S., Chen, Y., Kim, I., Kim, M.: Analysis of hyperspectral fluorescence images for poultry skin tumor inspection. Appl. Opt. 43(4), 824–833 (2004)
Kumar, S., Gosh, J., Crawford, M.M.: Best-bases feature extraction algorithm for classification of hyperspectral data. IEEE Trans Geosci Remote Sens., 39(7), 1368–1379 (2001)
Landgrebe, T., Paclík, P., Tax, D., Verzakov, S., Duin, R.: Cost-based classifier evaluation for imbalanced problems. In: Fred, A., Caelli, T., Duin, R., Campilho, A., de Ridder, D. (eds.) Structural, Syntactic, and Statistical Pattern Recognition, Proc. SSSPR2004 (Lisbon, Portugal, August 2004), LNCS, vol. 3138, pp. 762–770. Springer, Berlin Heidelberg New York (2004)
Lawrence, K., Windham, W., Park, B.R.J.B.: Hyperspectral imaging for poultry contaminant detection. NIR News 12(5) (2001)
Leemans, V., Destain, M.: A real-time grading method of apples based on features extracted from defects. J. Food Eng. 61, 83–89 (2004)
Leitner, R., Mairer, H., Kercek, A.: Real-time classification of polymers with nir spectral imaging and blob analysis. Real-Time Imaging 9(4), 245–251 (2003)
Noordam, J.: Chemometrics in multispectral imaging for quality inspection of postharvest products. Ph.D. thesis, Radboud University Nijmegen (2005)
Paclík, P., Duin, R.P.W.: Dissimilarity-based classification of spectra: computational issues. Real-Time Imaging 9(4), 237–244 (2003)
Paclík, P., Duin, R.P.W.: Designing multi-modal classifiers of spectra: a study on industrial sorting application. In: accepted for the 2nd Spectral Imaging workshop, Villach, Austria (2005)
Paclík, P., Tax, D., Verzakov, S., Duin, R.: Simplifying the model-based classifiers for multi-modal problems in classification of spectra. In: Nicolls, F. (eds.) Proc.of the Sixteenth Annual Symposium of the Pattern Recognition Association of South Africa, pp. 33–38. PRASA, Cape Town, RSA (2005). Langebaan, South Africa
Paclík, P., Verzakov, S., Duin, R.P.W.: Hypertools: the toolbox for spectral image analysis. Tech. rep., Pattern Recognition Group, TU Delft, The Netherlands (2003)
Paclík, P., Verzakov, S., Duin, R.P.W.: Improving the maximum-likelihood co-occurrence classifier: a study on classification of inhomogeneous rock images. In: Kalviainen, A.K.H., Parkinen, J., (eds.) Image Analysis—SCIA 2005 (Proc. 14th Scandinavian Conf. Joensuu, Finland, June 19–July 22), Lecture Notes in Computer Science, vol. 3540, pp. 998–1008. Springer, Berlin Heidelberg New York (2005)
Polder, G., van der Heijden, G., Young, I.: Tomato sorting using independent component analysis on specral images. Real-Time Imaging 9(4), 253–259 (2003)
Tatzer, P., Wolf, M., Panner, T.: Industrial application for inline material sorting using hyperspectral imaging in the nir range. Real-Time Imaging 11, 99–107 (2005)
Wold, S., Sjostrom, M.: SIMCA: a method for analysing chemical data in terms of similarity and analogy. ACS Symp. Series 52 (1977)
Yuhas, R.H., Goetz, A.F.H., Boardman, J.W.: Discrimination among semiarid landscape endmembers using the spectral angle mapper (SAM) algorithm. In: Summaries of the Third Annual JPL Airborne Geoscience Workshop, pp. 147–149. Jet Propulsion Laboratory, Pasadena, CA (1992)
Acknowledgments
The authors would like to thank Sergey Verzakov and Carmen Lai for fruitful comments on the manuscript. This research is/was supported by the Technology Foundation STW, applied science division of NWO and the technology program of the Ministry of Economic Affairs.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Paclík, P., Leitner, R. & Duin, R.P.W. A study on design of object sorting algorithms in the industrial application using hyperspectral imaging. J Real-Time Image Proc 1, 101–108 (2006). https://doi.org/10.1007/s11554-006-0018-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11554-006-0018-5