|
| This Article | ||
| ||
| Share | ||
| Bibliographic References | ||
| Add to: | ||
 Digg  Furl  Spurl  Blink  Simpy  Google  Del.icio.us  Y!MyWeb | ||
| Search | ||
| ||
| ASCII Text | x | ||
| Karen B. Sarachik, "The Effect of Gaussian Error in Object Recognition," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 4, pp. 289-301, April, 1997. | |||
| BibTex | x | ||
| @article{ 10.1109/34.587990, author = {Karen B. Sarachik}, title = {The Effect of Gaussian Error in Object Recognition}, journal ={IEEE Transactions on Pattern Analysis and Machine Intelligence}, volume = {19}, number = {4}, issn = {0162-8828}, year = {1997}, pages = {289-301}, doi = {http://doi.ieeecomputersociety.org/10.1109/34.587990}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, } | |||
| RefWorks Procite/RefMan/Endnote | x | ||
| TY - JOUR JO - IEEE Transactions on Pattern Analysis and Machine Intelligence TI - The Effect of Gaussian Error in Object Recognition IS - 4 SN - 0162-8828 SP289 EP301 EPD - 289-301 A1 - Karen B. Sarachik, PY - 1997 KW - Gaussian error models KW - object recognition KW - error analysis. VL - 19 JA - IEEE Transactions on Pattern Analysis and Machine Intelligence ER - | |||
Abstract—In model based recognition, the goal is to locate an instance of one or more known objects in an image. The problem is compounded in real images by the presence of clutter, occlusion, and sensor error, which can lead to "false negatives," failures to recognize the presence of the object, and "false positives," in which the algorithm incorrectly identifies an occurrence of the object. The probability of either event is affected by parameters within the recognition algorithm, which are almost always chosen in an ad-hoc fashion. The effect of the parameter values on the likelihood that the recognition algorithm will make a mistake are usually not understood explicitly.
To address the problem, we explicitly model the noise that occurs in the image. In a typical recognition algorithm, hypotheses about the position of the object are tested against the evidence in the image, and an overall score is assigned to each hypothesis. We use a statistical model to determine what score a correct or incorrect hypothesis is likely to have, and use standard binary hypothesis testing techniques to distinguish correct from incorrect hypotheses. Using this approach, we can compare algorithms and noise models, and automatically choose values for internal system thresholds to minimize the probability of making a mistake.
[1] T. Alter and W. Grimson, "Fast and Robust 3D Recognition by Alignment," Proc. Fourth Int'l Conf. Computer Vision, May 1993.
[2] H.S. Baird, Model Based Image Matching Using Location, ACM Distinguished Dissertation. Cambridge, Mass: MIT Press, 1984.
[3] T.A. Cass, "Feature Matching for Object Localization in the Presence of Uncertainty," Technical Report 1,133, MIT AI Laboratory, 1990.
[4] M. Costa, R. Haralick, and L. Shapiro, "Optimal Affine-Invariant Point Matching," Proc. Sixth Israeli Conf. Artificial Intelligence, pp. 35-61, 1990.
[5] M.A. Fischler and R.C. Bolles, "Random Sample Consensus: A Paradigm for Model Fitting With Applications to Image Analysis and Automated Cartography," Technical Report 213, SRI Int'l, Mar. 1980.
[6] W. Grimson, D. Huttenlocher, and T. Alter, "Recognizing 3D Objects from 2D Images: An Error Analysis," Technical Report 1,362, MIT AI Lab, 1992.
[7] W. Grimson, D. Huttenlocher, and D. Jacobs, "Affine Matching With Bounded Sensor Error: A Study of Geometric Hashing and Alignment," Technical Report 1,250, MIT AI Lab, 1991.
[8] W. Grimson, D. Huttenlocher, and D. Jacobs, "A Study of Affine Matching With Bounded Sensor Error," Int'l J. Computer Vision, vol. 13, no. 1, 1994.
[9] D.P. Huttenlocher, "Three-Dimensional Recognition of Solid Objects from a TwoDimensional Image," Technical Report 1,045, MIT AI Lab, 1988.
[10] D.W. Jacobs, "Recognizing 3D Objects Using 2D Images," Technical Report 1,416, MIT AI Lab, 1992.
[11] Y. Lamdan and H.J. Wolfson, “On the Error Analysis of‘Geometric Hashing’,” Proc. IEEE. Conf. Computer Vision and Pattern Recognition, pp. 22-27, 1991.
[12] I. Rigoutsos and R. Hummel, "Robust Similarity Invariant Matching in the Presence of Noise," Proc. Eighth Israeli Conf. Artificial Intelligence and Computer Vision, 1991.
[13] K. Sarachik, "An Analysis of the Effect of Gaussian Error in Object Recognition," Technical Report 1,469, MIT AI Lab, 1994.
[14] K.B. Sarachik and W. Grimson, "Gaussian Error Models for Object Recognition," CVPR, pp. 400-406, 1993.
[15] F.C.-D. Tsai, "A Probabilistic Approach to Geometric Hashing Using Line Features," Technical Report 640, Robotics Research Laboratory, New York Univ., June 1993.
[16] H.L. Van Trees, Detection, Estimation, and Modulation Theory, vol. I: "Detection, Estimation, and Linear Modulation Theory," chapter 2. John Wiley and Sons, 1968.
[17] W. Wells, "Statistical Object Recognition," Technical Report 1,395, MIT AI Lab, 1992.