Rotation- and scale-independent pattern recognition through optimization Export

@article{citeulike:2905935, abstract = {This paper presents a two-dimensional pattern recognition paradigm, which is independent of rotation and scale. It uses a Hopfield neural network with a configuration similar to that employed for the traveling salesman problem (TSP). Enhancements to the original Hopfield design are included which are based on the eigenvalues of the connection matrix and on modifications suggested by simulations. The goal of the method is to determine the underlying linear transformation between a test image and an observed input pattern by minimizing the energy of a Hopfield network. This procedure is easy to implement and is well suited for sparse binary images, e.g. skeletalized representations or the output from vertex detection algorithms. In this work the effects of noise are studied both analytically and through Monte Carlo techniques yielding some useful performance guidelines and greatly increasing the utility of the method. Experiments show that this technique performs successfully on a variety of rotated and scaled images and is robust against additive noise.}, author = {Hemminger, Thomas L. and Pomalaz-Raez, Carlos A.}, citeulike-article-id = {2905935}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/0031-3203(95)00100-X}, citeulike-linkout-1 = {http://www.sciencedirect.com/science/article/B6V14-3VSCPJ7-F/1/ee881ebe0b2e7f0fffd48bb152a8339a}, comment = {Summary: -objects: black and white images of patterns or images with extracted salient points; -features: black pixels in the image -classifier: NN Overview: In this paper, a method to classify patterns is presented. The algorithm is useful for recognition, as in OCR or printed circuit boards. It works by minimizing (optimizing) the "energy" needed to transform the input pattern to one of the possible output patterns. }, doi = {10.1016/0031-3203(95)00100-X}, journal = {Pattern Recognition}, keywords = {classification, optimization, orientation-invariant, pattern, scale-invariant}, month = {March}, number = {3}, pages = {487--495}, posted-at = {2008-06-18 19:13:21}, priority = {0}, title = {Rotation- and scale-independent pattern recognition through optimization}, url = {http://dx.doi.org/10.1016/0031-3203(95)00100-X}, volume = {29}, year = {1996} }

TY - JOUR ID - citeulike:2905935 L3 - citeulike-article-id:2905935 N2 - This paper presents a two-dimensional pattern recognition paradigm, which is independent of rotation and scale. It uses a Hopfield neural network with a configuration similar to that employed for the traveling salesman problem (TSP). Enhancements to the original Hopfield design are included which are based on the eigenvalues of the connection matrix and on modifications suggested by simulations. The goal of the method is to determine the underlying linear transformation between a test image and an observed input pattern by minimizing the energy of a Hopfield network. This procedure is easy to implement and is well suited for sparse binary images, e.g. skeletalized representations or the output from vertex detection algorithms. In this work the effects of noise are studied both analytically and through Monte Carlo techniques yielding some useful performance guidelines and greatly increasing the utility of the method. Experiments show that this technique performs successfully on a variety of rotated and scaled images and is robust against additive noise. IS - 3 JF - Pattern Recognition EP - 495 TI - Rotation- and scale-independent pattern recognition through optimization VL - 29 SP - 487 KW - classification KW - optimization KW - orientation-invariant KW - pattern KW - scale-invariant N1 - Summary: -objects: black and white images of patterns or images with extracted salient points; -features: black pixels in the image -classifier: NN Overview: In this paper, a method to classify patterns is presented. The algorithm is useful for recognition, as in OCR or printed circuit boards. It works by minimizing (optimizing) the "energy" needed to transform the input pattern to one of the possible output patterns. AU - Hemminger, Thomas AU - Pomalaz-Raez, Carlos PY - 1996/03// UR - http://dx.doi.org/10.1016/0031-3203(95)00100-X ER -