Artificial intelligence

Abstract: The article is devoted to a problem of acceleration of objects detection process on the images, the multiscale scanning based on use. For the solution of this task it is offered to use preliminary processing of candidates with use of integrated characteristics which is realized as the first stage of the classifiers cascade of the mixed type. As a test task the problem of detection on digital images of cars registration plates is chosen. The received cascade of classifiers allowed to increase computational performance of images processing at detection of registration plates by 1.6 times in comparison with the cascade of linear classifiers received by means of AdaBoost algorithm. Results of the made experiments can be extended to problems of search of other objects on images.

Keywords: image recognition, object detection, image analysis, image classification

References:

1. Murygin K.V. Obnaruzhenie ob’ektov na izobrazhenii na osnove kaskada klassifikatorov // Iskusstvennyj intellekt. 2007. - No2. - S.104-108.
2. Paul Viola and Michael J. Jones. Robust real-time object detection. //In Proc. of IEEE Workshop on Statistical and ComputationalTheories ofVision, 2001.
3. Bradski G., Kaehler A., Pisarevsky V. Learning-based computer vision with Intel’s open source computer vision library //Intel Technology Journal. http://develiper.intel.com/technology/itj/index.htm.
4. Robert E. Schapire, Freund Y. Bartlett P, Lee W.S. Boosting the margin: A new explanation for the effectiveness of voting methods // In Proc. Of the Fourteenth International Conference on Machine Learning, 1997.
5. Murygin K.V. Osobennosti realizacii algoritma AdaBoost dlja obnaruzhenija ob’ektov na izobrazhenijah // Iskusstvennyj intellekt. 2009. - No3. - S. 573-581.
6. Murygin K.V. Kombinirovanie binarnyh svojstv v vide MKV-klassifikatora // Iskusstvennyj intellekt. 2010. – No 1. – S. 108-113.
7. Murygin K.V. Predvaritel'naja obrabotka kandidatov pri obnaruzhenii avtomobil'nyh nomerov na izobrazhenijah // Iskusstvennyj intellekt. – 2013. – No 2. – S. 32-37.