0434521 20240103204954.920141124235959.9 84924690353 000351213400007 10.1109/TPAMI.2014.2353644 17 s. P cav_un_epca*02567250162-8828374 (2015)786-802IEEE Computer Society Blurred image N-fold rotation symmetry projection operators image moments moment invariants blur invariants object recognition cav_un_auth*0101087 Zpracování obrazové informace Department of Image Processing ZOI ZOI Department of Image Processing Flusser Jan UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101203 Zpracování obrazové informace Department of Image Processing ZOI ZOI Department of Image Processing Suk Tomáš UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101071 Zpracování obrazové informace Department of Image Processing ZOI ZOI Department of Image Processing Boldyš Jiří UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101238 Zpracování obrazové informace Department of Image Processing ZOI ZOI Department of Image Processing Zitová Barbara UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. http://library.utia.cas.cz/separaty/2014/ZOI/flusser-0434521.pdf cav_un_auth*0292734 GA13-29225S GA ČR cav_un_auth*0273618 GAP103/11/1552 GA ČR In this paper we introduce a new theory of blur invariants. Blur invariants are image features which preserve their values if the image is convolved by a point-spread function (PSF) of a certain class. We present the invariants to convolution with an arbitrary N-fold symmetric PSF, both in Fourier and image domain. We introduce a notion of a primordial image as a canonical form of all blur-equivalent images. It is defined in spectral domain by means of projection operators. We prove that the moments of the primordial image are invariant to blur and we derive recursive formulae for their direct computation without actually constructing the primordial image. We further prove they form a complete set of invariants and show how to extent their invariance also to translation, rotation and scaling. We illustrate by simulated and real-data experiments their invariance and recognition power. Potential applications of this method are wherever one wants to recognize objects on blurred images. JD 2017 4 4 A hod 4ah 20231122140557.6 RVO:67985556 http://hdl.handle.net/11104/0239389 1 Department of Image Processing UTIA-B 10200 COMPUTER SCIENCE, THEORY & METHODS S COMPUTERSCIENCE.ARTIFICIALINTELLIGENCE|ENGINEERING.ELECTRICAL&ELECTRONIC 8.693 0.731 999.9 0.0612 4.15 31757 6.202 Q1 5.860 193 Q1 99.622 98.2 Q1* Q1* 98.249 2015 Soubory v repozitáři: flusser-0434521.pdf 84924690353 SCOPUS 000351213400007 WOS cav_un_epca*0256725 IEEE Transactions on Pattern Analysis and Machine Intelligence 0162-8828 1939-3539 Roč. 37 č. 4 2015 786 802 IEEE Computer Society