Editing Pixelization (section)

==Alternative techniques==
[[File:Umm Rasas Fisherman.JPG|thumb|The mosaic tiles of this human figure in the Saint Stephen Church in [[Umm ar-Rasas]] in Jordan have been rearranged by [[Aniconism|aniconists]].]]
A black rectangular or square box (known as [[censor bars]]) may be used to occlude parts of images completely (for example, a black bar covering the eyes instead of the entire face being pixelized). Censor bars were extensively used as a graphic device in the January 2012 [[protests against SOPA and PIPA]].

A drawback of pixelization is that any differences between the large pixels can be exploited in moving images to reconstruct the original, unpixelized image;<ref>{{cite journal |last1=Newton |first1=EM |last2=Sweeney |first2=L |last3=Malin |first3=B |title=Preserving privacy by de-identifying face images |journal=IEEE Transactions on Knowledge and Data Engineering |date=Jan 2005 |volume=17 |issue=2 |pages=232–243 |doi=10.1109/TKDE.2005.32 |s2cid=1813467 |url=https://dataprivacylab.org/dataprivacy/projects/video/paper.pdf}}</ref> squinting at a pixelized, moving image can sometimes achieve a similar result. In both cases, integration of the large pixels over time allows smaller, more accurate pixels to be constructed in a still image result. Completely obscuring the censored area with pixels of a constant color or pixels of random colors escapes this drawback but can be more aesthetically jarring.

An additional drawback, when pixelization is used to reduce the repulsing, disturbing or, more generally shocking, aspect of an image, is that all information contained in the pixelized area is lost for the audience.<ref name="BesançonSemmo2019">{{cite journal|last1=Besançon|first1=Lonni|last2=Semmo|first2=Amir|last3=Biau|first3=David|last4=Frachet|first4=Bruno|last5=Pineau|first5=Virginie|last6=Sariali|first6=El Hadi|last7=Soubeyrand|first7=Marc|last8=Taouachi|first8=Rabah|last9=Isenberg|first9=Tobias|last10=Dragicevic|first10=Pierre|title=Reducing Affective Responses to Surgical Images and Videos Through Stylization|journal=Computer Graphics Forum|volume=39|issue=1|year=2019|pages=462–483|issn=0167-7055|doi=10.1111/cgf.13886|doi-access=free}}</ref> Other visual processing techniques can help reduce the shocking aspect of images or videos while preserving most of the information of the media.<ref name="BesançonSemmo2019"/><ref name="BesançonSemmo2018">{{cite book|last1=Besançon|first1=Lonni|last2=Semmo|first2=Amir|last3=Biau|first3=David|last4=Frachet|first4=Bruno|last5=Pineau|first5=Virginie|last6=Sariali|first6=El Hadi|last7=Taouachi|first7=Rabah|last8=Isenberg|first8=Tobias|last9=Dragicevic|first9=Pierre|title=Proceedings of the Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering|chapter=Reducing affective responses to surgical images through color manipulation and stylization|year=2018|pages=1–13|doi=10.1145/3229147.3229158|isbn=9781450358927|s2cid=51928057|url=https://hal.inria.fr/hal-01795744/file/Besancon_2018_RAR.pdf}}</ref>