Frank Agbat Posted October 17, 2006 Share Posted October 17, 2006 Often researchers must use what is at hand, which is most oftena halftone. Few researchers have access to originals. When doing this, researcher must be aware that halftones may introduce artifacts not present in the original. Most prominent false conclusions are the early studies of Raymond Marcus, who thought he saw a man with a gun in a Moorman halftone, below. The image does NOT appear in ANY continuous tone Moorman copy. Jack Jack, Spot on! Artifacts, plus the way the human brain interprets shapes and patterns will lead people down the path of doom (so to speak). The example you shared is a classic case of this! Another point that is relevant to mention here is that various digital compression techniques used to save images also introduce loss and artifacting. Although the resulting images are much smaller in size (and thus more portable). JPEG, for example, is one (of many) digital image formats that introduce loss and artifacting. I've seen any number of well-intended attempts at analysis that start with a 100K Jpeg that has been digitally "zoomed", etc, etc. The various "faces" and other "details" that people "see" are probably nothing but jpeg blocking and aliasing that happen to form a 3-dot pattern (ie. a "face"). Link to comment Share on other sites More sharing options...
John Dolva Posted October 18, 2006 Author Share Posted October 18, 2006 Often researchers must use what is at hand, which is most often a halftone. Few researchers have access to originals. When doing this, researcher must be aware that halftones may introduce artifacts not present in the original. Most prominent false conclusions are the early studies of Raymond Marcus, who thought he saw a man with a gun in a Moorman halftone, below. The image does NOT appear in ANY continuous tone Moorman copy. Jack Jack, Spot on! Artifacts, plus the way the human brain interprets shapes and patterns will lead people down the path of doom (so to speak). The example you shared is a classic case of this! Another point that is relevant to mention here is that various digital compression techniques used to save images also introduce loss and artifacting. Although the resulting images are much smaller in size (and thus more portable). JPEG, for example, is one (of many) digital image formats that introduce loss and artifacting. I've seen any number of well-intended attempts at analysis that start with a 100K Jpeg that has been digitally "zoomed", etc, etc. The various "faces" and other "details" that people "see" are probably nothing but jpeg blocking and aliasing that happen to form a 3-dot pattern (ie. a "face"). This is very important. Even when one has a result one is happy with, any* resizing and rotation introduces artefacts, or alters the data. Some algorithms much more so than others. Image analyzer http://meesoft.logicnet.dk/Analyzer/help.htm is a good tool for this It is very possible to find Lucy in the Sky with a cupboard full of jewels in many of the grainy images if one tries hard enough. *except a "pixel resize" Link to comment Share on other sites More sharing options...
John Dolva Posted October 20, 2006 Author Share Posted October 20, 2006 (edited) as example here is a small portion of a simple image resized with four different algorithms. The pixel resizing is the only 'true' one. The other three has various degrees of alteration. Again a portion of this is resized and manipulated with equalisations smoothings sharpening etc and then a portion of that selected and colorised to show a policeman. I'm not saying that this is what happened with regards to 'badgeman'. Rather this is an exaggeration in order to higlight a necessary consideration with regards to computer assisted image manipulations, even where, as in this case, it is not a matter of starting off with a 'faulty' image (scanned grainy etc) but just the usual processes people use to enlarge with computers. Edited October 20, 2006 by John Dolva Link to comment Share on other sites More sharing options...
Jack White Posted October 20, 2006 Share Posted October 20, 2006 (edited) as example here is a small portion of a simple image resized with four different algorithms. The pixel resizing is the only 'true' one. The other three has various degrees of alteration. Again a portion of this is resized and manipulated with equalisations smoothings sharpening etc and then a portion of that selected and colorised to show a policeman. I'm not saying that this is what happened with regards to 'badgeman'. Rather this is an exaggeration in order to higlight a necessary consideration with regards to computer assisted image manipulations, even where, as in this case, it is not a matter of starting off with a 'faulty' image (scanned grainy etc) but just the usual processes people use to enlarge with computers. John...you seem to be under the mistaken idea that Badgeman was "discovered" by COMPUTER. Wrong. Your premise is faulty. ALL Badgeman work was done in pre-computer days, the old fashioned way: PHOTOGRAPHY (camera, film, paper, developer, darkroom). This was TWELVE YEARS BEFORE I HAD A COMPUTER and 14 years before I was on the internet. There was NEVER a computer image used, and the original was NOT "FAULTY", BUT PERFECTLY CLEAR. Jack Edited October 20, 2006 by Jack White Link to comment Share on other sites More sharing options...
John Dolva Posted October 20, 2006 Author Share Posted October 20, 2006 Jack, I'm aware of that, and therefore I wrote: "I'm not saying that this is what happened with regards to 'badgeman'." and "Rather this is an exaggeration in order to higlight a necessary consideration with regards to computer assisted image manipulations, even where, as in this case, it is not a matter of starting off with a 'faulty' image (scanned grainy etc) but just the usual processes people use to enlarge with computers." IOW when one does use computers on should be aware of this. I realise you did use optical techniques and that is a different thing. These days, many don't do that and it's possible to make mistakes. I've actually decided I cannot comment, for these reasons, on Badgeman because I don't have anything but what I can get digitally. I had hoped to find it to be different. Link to comment Share on other sites More sharing options...
Jack White Posted October 20, 2006 Share Posted October 20, 2006 Jack, I'm aware of that, and therefore I wrote:"I'm not saying that this is what happened with regards to 'badgeman'." and "Rather this is an exaggeration in order to higlight a necessary consideration with regards to computer assisted image manipulations, even where, as in this case, it is not a matter of starting off with a 'faulty' image (scanned grainy etc) but just the usual processes people use to enlarge with computers." IOW when one does use computers on should be aware of this. I realise you did use optical techniques and that is a different thing. These days, many don't do that and it's possible to make mistakes. I've actually decided I cannot comment, for these reasons, on Badgeman because I don't have anything but what I can get digitally. I had hoped to find it to be different. You are correct, but your original wording did not reflect that. Jack Link to comment Share on other sites More sharing options...
John Dolva Posted October 20, 2006 Author Share Posted October 20, 2006 I'm not sure in what sense you mean Jack. I didn't change the words, just elaborated. I had hoped to be able to use imagery presented to support or not the 'Badgeman'. I am coming to the conclusion that I cannot one way or the other, and as others are doing so using computers, I would question their conclusions. This is separate from your work. As far as your work goes, until someone repeats it in the same way, it stands apart. And there only the arguments re. lens resolution, media, dust etc are relevant. So I excuse myself and leave that one to the relevant experts. Yourself and others. Link to comment Share on other sites More sharing options...
Jack White Posted October 20, 2006 Share Posted October 20, 2006 I'm not sure in what sense you mean Jack. I didn't change the words, just elaborated. I had hoped to be able to use imagery presented to support or not the 'Badgeman'. I am coming to the conclusion that I cannot one way or the other, and as others are doing so using computers, I would question their conclusions. This is separate from your work. As far as your work goes, until someone repeats it in the same way, it stands apart.And there only the arguments re. lens resolution, media, dust etc are relevant. So I excuse myself and leave that one to the relevant experts. Yourself and others. Thanks, John. I should have said I MISINTERPRETED what you wrote. Sorry that I was not clear. Jack Link to comment Share on other sites More sharing options...
John Dolva Posted October 20, 2006 Author Share Posted October 20, 2006 (edited) No probs, Jack, thank's for the exchange. I've found this thread input most helpful. I guess as always I'm looking for that elusive threshold on one side of which lies specultion and the other, certainty. I think it might be possible to develop some sort of more precice yardstick as far as computer manipulated imagery goes. ie any given image enlarged beyond Xpower or x# pixel grid will necessarily be beyond 'trust'. I remember in the late 70's at uni when a lot of images were disrtributed as characterised printouts. I remember someone wrote something on image recognition and the number of characters that can represent a recognisable face. Interestingly it used Kennedys face. From memory it was of a grid, something in the order of tenxten or twentyxtwenty characters. Edited October 20, 2006 by John Dolva Link to comment Share on other sites More sharing options...
John Dolva Posted October 20, 2006 Author Share Posted October 20, 2006 (edited) NNNHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH NH©m8805000050000000088888888888AAAA880000008AQQA808QQ N©m5+*(!!;;;!;;*+$20AQQQAmEmmmmmEE#mA0$=((((=+%%+*(*%2 N©A$*!!:':::,;%$588mQEEE#E#EEAAAAA8888085=;;(===*;!;=$ N©A%(,:'':'!%$%208Q8AmE##EEmQ88008A8AAQAm8=((***(;,;=$ N©A%;::'.';(%%200m##AQQmEEEmmmQ88A808QmE##A5+**((!,;*$ N©A+!'''',**(*2mmEEEAQAmEEEEQAA88AQAAQmm#©©©E$*((,,,*% N#8+,''.'(((*%0A#QQE##©©©EmEmQQQmQmEEEA0E####A*;;,,,(% N©8=,'..:;+=(%$5E©©©©HHHHHH#mmE###EEm0%$A8AQmE=(!:::;% N#8=:'..'!(((%20AmE©©©©HH©©###EQA05%*;!(%$58AA$;,:::;+ N#8=,'.''!(=;:,!,($0000052$%++=*;,,,,,!;%0AmQ0$;::::!= N#8=,'..':,(!..,;*=%++==**(**(;;,,,,,!!;*0QQA0%!':''!* N#8=,'...':!;:'!;*==+%+=+=*((((!::':,!!;=0mmA0=,::'',* N#8=,'...':,;!',((*====*=**((;;!,,:::,,,(8EQQ0(:':'',* H#8=,'...':!:...!*=%$$%+$+*;**+$22%*;!!,!+Qm85(''''',* H#8=,''..:,,...:=$2228QmQ0(,*58805+((**!,!*AA$+!'''',* N#8=,'..'=;,..:,208mA%E©m=!:!$AQ8Qm52%(;,,;02((!..'',* H©8=,'..:%;*..!(*+55$$mE((;!;;+805%=(!,,,,;=%+(,.''',* N#8=,''.'=5Q'.:*$%50AA2(!(;;(;;,(=*!:'':,!;(;=(,.''',* N#8=!''':!%A,'+%%%$2*(:](;(((!,,,!,:!;;;**!;!'.''',* N#8=,''',;;,,,;*+%%20m5;($+*=*(*=(;;;;;(((**!!:..'''!* N#8=!:'.:!:::,;*=%28mE02080$52*!;==(;;;;((**!:'.'''';* N#8+!:'':,:::,!((=$08%=+%$+$*;;;!!;(;(((((((,'..'''';= N©8+!:'':,:':,;(;*+$2%$$%%===****++*((((**=;:'...'':;= N©8+!:''::''':;;(=%5m#A0AAA05$++++52=***=+*!:'...''= N©8+!:''::'':',;*=%8Q8%%5222$%==**=$+++%$5+(,'...'',(= N©8+!:'',,!(==(=++%52$$2088052+*((*=%%$$$QmmQA82+=*(*+ H©8%*+28mm©©Q5522++%==++==++=(;;;(*$55$+$mQQQmEEEEEEmm HH©©HHH©HHHm00A528($$++%$2$$2%===+582$%%0mQQQQmmmEEEEE NNNNHHHNNH###Q050m$(+2520QAAQQA800522%%!QmmQQQQQmE##EE NNNNNNNNN©©©©EmAAEA'!502255QmA00002%$%,(QmmmQmQmmmE#EE NNNNNNNNHHHH©#mmEm©$.,%58A08Q050052$(..2mmmmmmmmmmmEEE NNNNNNNH©HNNH#EEEE©©*..,+0mmmAAA8$;' !EmmmmmmmmmmmmEE NNNNNNH©©NNNH©©#E##©©='':::!=55%, .2#mmEmmmEmmmmmmm NNNHHHH©©NNNNH©####©H©$,:''..:''. (#mmmmmmmmmmmmmmE NNHHH©HHHHHHN©©©##E#©©E2(:.'::::]. :m#EEEmmmmmmmmmmmm Image Analyzer characterise feature, courier new size 7 and for interest sake here is the above letter arrangment snapshotted, reduced in size and blurred and histogram equalised Edited October 20, 2006 by John Dolva Link to comment Share on other sites More sharing options...
Jack White Posted October 20, 2006 Share Posted October 20, 2006 Verrry interesting! Jack Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now