15 ms·

Finding Lena Forsen, the Patron Saint of JPEGs

242 points by lelf 8y ago

dav 8y ago
Did she actually still have the same hat and scarf? I didn't notice that mentioned in the text.
- saagarjha 8y ago
  The new one looks slightly different: it has a different texture on the underside of the brim. It’s likely that the original hat belonged to Playboy.
FartyMcFarter 8y ago
> The small army of women who had worked as so-called computers during the first half of the 20th century were leaving the tech industry in droves, as computing went from being perceived as a form of menial labor to a more cerebral, masculine pursuit. Isn't this rather misleading? Obviously there were women in early computing who were very influential. But weren't the "human computers" in fact performing tedious and repetitive calculations, and not merely "perceived" as such?
- regularfry 8y ago
  No, it's not misleading. The change was in perception, not in the nature of the task. Yes, there were women who did rote calculations, but those weren't the roles influential to how the role evolved.
  - erik_seaberg 8y ago
    "Programmer" used to be the job of turning prose or flowcharts from a "systems analyst" into a deck of machine instructions. The uncreative grunt work making up half of the profession got completely automated away.
    - analog31 8y ago
      I know a person who had that kind of job, and eventually started fixing bugs in the code because it was easier than tracking down the analyst. She gradually morphed into being a full fledged programmer in the contemporary sense, now nearing retirement.
  - FartyMcFarter 8y ago
    But the quoted sentence explicitly mentions an "army of women" doing human computer jobs.
ChuckMcM 8y ago
Oh that is fun, I was one of the operators at the USC Image Processing Institute who did scanning, ran the systems (initially a the HP2100, then a PDP 11/40, and later a PDP 11/55t. We did lots of tours and we had a number of graduate students working on various issues. One of the things we did was to provide the 512 x 512 set of reference images on 1600BPI magnetic tape to researchers. Each image was stored as a red image, green image, and blue image (8 bits per pixel for 24 bits total). In addition to Lena there were other standard images that involved repeating texture, lots of lines and edges, and variable frequency details. By using the standard images, two algorithms could be compared using their output on the same standard image. This helped researchers in the field compare their work with other published studies.
- saagarjha 8y ago
  The images (well, most of them) are online: http://sipi.usc.edu/database/ http://sipi.usc.edu/database/
  - aboutruby 8y ago
    > Please note that we no longer distribute the following images that were previously available in our database: 4.2.04 (lena), 4.2.02 (tiffany), elaine.512, numbers.512 and testpat.1k. Although these images have played a significant role in the history of image processing, they no longer represent the best examples for future research.
    - misterdoubt 8y ago
      Hallelujah.
    - saagarjha 8y ago
      Yup, that’s what I was referring to. Kind of disappointing that they’ve taken down Lena :(
    - GuB-42 8y ago
      It there a reason why these images "no longer represent the best examples for future research"? How is Lena inferior to the other "female" images? Lena seems to offer a good mix of highly detailed textures, smooth gradients and sharp edges. Maybe there are copyright reasons but Playboy seems to be fine with the use that the Lena image is made of.
ralusek 8y ago
"In 1973, at the moment that her picture was being brought into the lab, there were hundreds if not thousands of women being pushed out,” said Marie Hicks, a historian of technology and author of Programmed Inequality.1 “All this happened for a reason. If they hadn’t used a Playboy centerfold, they almost certainly would have used another picture of a pretty white woman. The Playboy thing gets our attention, but really what it’s about is this world-building that’s gone on in computing from the beginning—it’s about building worlds for certain people and not for others." This is the thesis of the article, if you don't want to waste your time reading through the banal padding of a story around it.
- Walf 8y ago
  It's definitely pushing an agenda, with a whiff of product placement, too. It's a bit of an absurd conclusion they're drawing. Sure, continuing to use a centerfold picture is sexist, but the picture was only a tool and - unlike the article implies - there is no 'spirit' of Lena inside my family photos.
iamleppert 8y ago
What’s wrong with female beauty? It’s not like it’s porn or a nude image. I’m not even a straight male but I can still appreciate female beauty and I don’t think it’s sexist. The photo is more artistic than anything. I wonder if the same people who complain about this photo lodge similar complaints when they visit art museums that have female nudity in oil paintings or nude female sculptures? I know nudity and even a slight sexual undertone in anything upsets some people, but I don’t understand why. I think it says something more about the person than our culture. Sexuality is a core part of the human experience and to deny it is to remove that part of our humanity.
- Pharmakon 8y ago
  Personally I’m not bothered, and while the image isn’t pornographic, it is from Playboy which does make it at least porn adjacent. I see it’s use less as a matter of offensiveness or some headfuck of “intersectionality” and more as just immature and unprofessional. Sexuality is great, it’s everywhere, and I think it has literally nothing to do with this issue. I can understand not wanting a Playboy model to be the standard image in a professional setting, especially a setting that is overwhelmingly full of guys.
  - qubex 8y ago
    I’m a nudist (meaning I have no problem being naked or seeing people without clothes on when it is comfortable, publicly or privately). What on earth is wrong with a nude photograph? What does it say about humankind that we deny our baseline state?
    - Pharmakon 8y ago
      Fair play to you on the nudism. My question would be whether you’d show up starkers at work? If not, then I’m not sure how it applies.
      - qubex 8y ago
        No I wouldn’t. Firstly because shocking unexpected people is streaking and/or (in some jurisdictions at least) some kind of offence. Secondly because I’m willing to see people undressed doesn’t mean it’s universal (everybody has there preferences, prejudices, and ingrained values). Thirdly because I’m the managing director and enthusiastic or not or not one of my duties is to set the tone for the workplace by example not by demanding that others do what I tell them to but by leading by example, and that includes conservative dress. Then again I’m in Italy where somewhat different social norms apply: people ask each other out on dates after they meet in the workplace and even if it’s unappreciated nobody dreams of kicking up a stink for sexual harassment because they consider it a normal form of human interaction. It’s also normal to compliment people on their looks and choice of attire. Both genders do it. Nobody gets offended. It’s more like “yeah, I put a ton of effort into choosing and matching this outfit this morning, and nobody noticed?! WTF??”
xellisx 8y ago
I thought the image was used to test formats well before JPEG.
Jun8 8y ago
In addition to lenna old timer image processing researchers would also have fond (depending whether your compression or edge detection algorithm worked correctly or not) memories of girl, cameraman, and barbara images as well as the infamous baboon (https://goo.gl/images/4Js7YH https://goo.gl/images/4Js7YH). I remember doing my senior thesis on 2D LPC in 1990 and since we did not know how to print images, in our report we just printed the pixel values as numbers. Ahh, the good old times.
- toolslive 8y ago
  infamous mandrill, I think it is. Anyway, that picture in full colour has a definite wow-factor. https://www.researchgate.net/figure/Original-standard-test-image-of-Mandrill-also-known-as-Baboon_fig9_259521525 https://www.researchgate.net/figure/Original-standard-test-i...
jordigh 8y ago
I feel that whenever Lena is mentioned, we should also give a shoutout to the researchers who were fed up with her and decided to use Fabio instead. https://arxiv.org/abs/1202.6429 https://arxiv.org/abs/1202.6429 What do you guys (yes, guys) think about making Fabio the standard image instead?
- qubex 8y ago
  Well on her (the researcher), her (Lena), and him (Fabio). Well on all of them.
  - brokenmachine 8y ago
    You mean "good on them"?
- johndough 8y ago
  The Fabio image is not very useful wrt image compression: - grayscale (lenna image is rgb) - very odd 271x257 resolution (lenna image is square 512x512 which is a power of two) - just hair as fine detail features (lenna image has several: hat, feather boa and hair) https://4.bp.blogspot.com/-jUHXZ0jW4Lc/T1t3W0ex2BI/AAAAAAAAFeY/7TY0cTty1og/s400/Fabio.jpg https://4.bp.blogspot.com/-jUHXZ0jW4Lc/T1t3W0ex2BI/AAAAAAAAF... http://www.betapleatedchic.com/uploads/3/0/4/1/30413754/4907429_orig.gif http://www.betapleatedchic.com/uploads/3/0/4/1/30413754/4907... An optimal image would have the following properties: - different textures at different - scale - brightness - frequency - easily recognizable features - color - power of two resolution - no cultural issues - no licensing issues (not a scan of a magazine, globally recognized license, i.e. not "public domain") The problem with faces is that on the one hand, humans are able to tell easily if a feature was destroyed by an image compression algorithm, but on the other hand it's personally identifiable information and therefore comes with additional legal restrictions in many jurisdictions.
- craigsmansion 8y ago
  > we should also give a shoutout No, we shouldn't. Them using a different image in their research is perfectly acceptable--when one spends hours looking at something rendering, one shouldn't have to look at something one objects to--but referring to the research as a social statement diminishes the scientific value by framing it as a publicity stunt. Also, apart from using research time/money for licensing: "lena" has become shorthand for "face-female-with-various-patterns" and is a historical accident. "Fabio" will always be Fabio, a commercial model; a less than optimal choice for a scientific paper.
59nadir 8y ago
Lena Forsén* You'd think, having written a whole article about her, you'd be able to spell her name right. It's incredible to me how relaxed people in English speaking countries, particularly the US, are with removing and changing letters as they see fit. I remember seeing someone on Reddit, for example, who had named their kid Soren, after Søren Kierkegaard, completely ignoring that it's not even the same name. Do people realize that these letters are different looking for a reason?
- rococode 8y ago
  Isn't that just how language works? When a name doesn't suit your language's writing system it's natural to adapt it in a way that's more accessible. English simply doesn't have an acute accent. It's why people in English speaking countries are named Michael (or one of the 50+ alternatives) instead of the original מִיכָאֵל, or why BBC reports on Chinese President Xi Jinping instead of Chinese President 习近平.
  - jabberthemutt 8y ago
    It is also natural to be attracted to sweet pictures of the other sex. But this whole article pushes the mindset that this is not ok. So I would argue that ignorance of proper spellings (in the same alphabet! Your examples are misleading) marginalizes them.
    - int_19h 8y ago
      It's not the same alphabet. English alphabet is distinct from the Swedish alphabet, they just happen to use the same writing system.
      - jabberthemutt 8y ago
        Sorry, second language English here. I thought it was obvious what I meant. é is very close in appearance and sound to e compared to some Hebrew(?) or Chinese.
  - 8y ago
loosetypes 8y ago
I wasn't aware of the compared cases from the article such as Suzanne Vega, "The Mother of the MP3", or Susan Bennett, the namesake for Siri. But I have heard that a 3d model of Thom Yorke (Radiohead)'s head is the de facto standard for 3d printer trials.
- abricot 8y ago
  So it isn't Benchy?
5555624 8y ago
> Yet the model herself (right, now 67) remained mostly a mystery—until now. Not exactly, there is even a web site about it -- http://www.lenna.org/ http://www.lenna.org/ -- and both she and the image have Wikipedia entries.

242 points by lelf 8y ago

dav 8y ago

Did she actually still have the same hat and scarf? I didn't notice that mentioned in the text.

saagarjha 8y ago
The new one looks slightly different: it has a different texture on the underside of the brim. It’s likely that the original hat belonged to Playboy.

FartyMcFarter 8y ago

> The small army of women who had worked as so-called computers during the first half of the 20th century were leaving the tech industry in droves, as computing went from being perceived as a form of menial labor to a more cerebral, masculine pursuit. Isn't this rather misleading? Obviously there were women in early computing who were very influential. But weren't the "human computers" in fact performing tedious and repetitive calculations, and not merely "perceived" as such?

regularfry 8y ago
No, it's not misleading. The change was in perception, not in the nature of the task. Yes, there were women who did rote calculations, but those weren't the roles influential to how the role evolved.
- erik_seaberg 8y ago
  "Programmer" used to be the job of turning prose or flowcharts from a "systems analyst" into a deck of machine instructions. The uncreative grunt work making up half of the profession got completely automated away.
  - analog31 8y ago
    I know a person who had that kind of job, and eventually started fixing bugs in the code because it was easier than tracking down the analyst. She gradually morphed into being a full fledged programmer in the contemporary sense, now nearing retirement.
- FartyMcFarter 8y ago
  But the quoted sentence explicitly mentions an "army of women" doing human computer jobs.

ChuckMcM 8y ago

Oh that is fun, I was one of the operators at the USC Image Processing Institute who did scanning, ran the systems (initially a the HP2100, then a PDP 11/40, and later a PDP 11/55t. We did lots of tours and we had a number of graduate students working on various issues. One of the things we did was to provide the 512 x 512 set of reference images on 1600BPI magnetic tape to researchers. Each image was stored as a red image, green image, and blue image (8 bits per pixel for 24 bits total). In addition to Lena there were other standard images that involved repeating texture, lots of lines and edges, and variable frequency details. By using the standard images, two algorithms could be compared using their output on the same standard image. This helped researchers in the field compare their work with other published studies.

saagarjha 8y ago
The images (well, most of them) are online: http://sipi.usc.edu/database/ http://sipi.usc.edu/database/
- aboutruby 8y ago
  > Please note that we no longer distribute the following images that were previously available in our database: 4.2.04 (lena), 4.2.02 (tiffany), elaine.512, numbers.512 and testpat.1k. Although these images have played a significant role in the history of image processing, they no longer represent the best examples for future research.
  - misterdoubt 8y ago
    Hallelujah.
  - saagarjha 8y ago
    Yup, that’s what I was referring to. Kind of disappointing that they’ve taken down Lena :(
  - GuB-42 8y ago
    It there a reason why these images "no longer represent the best examples for future research"? How is Lena inferior to the other "female" images? Lena seems to offer a good mix of highly detailed textures, smooth gradients and sharp edges. Maybe there are copyright reasons but Playboy seems to be fine with the use that the Lena image is made of.

ralusek 8y ago

"In 1973, at the moment that her picture was being brought into the lab, there were hundreds if not thousands of women being pushed out,” said Marie Hicks, a historian of technology and author of Programmed Inequality.1 “All this happened for a reason. If they hadn’t used a Playboy centerfold, they almost certainly would have used another picture of a pretty white woman. The Playboy thing gets our attention, but really what it’s about is this world-building that’s gone on in computing from the beginning—it’s about building worlds for certain people and not for others." This is the thesis of the article, if you don't want to waste your time reading through the banal padding of a story around it.

Walf 8y ago
It's definitely pushing an agenda, with a whiff of product placement, too. It's a bit of an absurd conclusion they're drawing. Sure, continuing to use a centerfold picture is sexist, but the picture was only a tool and - unlike the article implies - there is no 'spirit' of Lena inside my family photos.

iamleppert 8y ago

What’s wrong with female beauty? It’s not like it’s porn or a nude image. I’m not even a straight male but I can still appreciate female beauty and I don’t think it’s sexist. The photo is more artistic than anything. I wonder if the same people who complain about this photo lodge similar complaints when they visit art museums that have female nudity in oil paintings or nude female sculptures? I know nudity and even a slight sexual undertone in anything upsets some people, but I don’t understand why. I think it says something more about the person than our culture. Sexuality is a core part of the human experience and to deny it is to remove that part of our humanity.

Pharmakon 8y ago
Personally I’m not bothered, and while the image isn’t pornographic, it is from Playboy which does make it at least porn adjacent. I see it’s use less as a matter of offensiveness or some headfuck of “intersectionality” and more as just immature and unprofessional. Sexuality is great, it’s everywhere, and I think it has literally nothing to do with this issue. I can understand not wanting a Playboy model to be the standard image in a professional setting, especially a setting that is overwhelmingly full of guys.
- qubex 8y ago
  I’m a nudist (meaning I have no problem being naked or seeing people without clothes on when it is comfortable, publicly or privately). What on earth is wrong with a nude photograph? What does it say about humankind that we deny our baseline state?
  - Pharmakon 8y ago
    Fair play to you on the nudism. My question would be whether you’d show up starkers at work? If not, then I’m not sure how it applies.
    - qubex 8y ago
      No I wouldn’t. Firstly because shocking unexpected people is streaking and/or (in some jurisdictions at least) some kind of offence. Secondly because I’m willing to see people undressed doesn’t mean it’s universal (everybody has there preferences, prejudices, and ingrained values). Thirdly because I’m the managing director and enthusiastic or not or not one of my duties is to set the tone for the workplace by example not by demanding that others do what I tell them to but by leading by example, and that includes conservative dress. Then again I’m in Italy where somewhat different social norms apply: people ask each other out on dates after they meet in the workplace and even if it’s unappreciated nobody dreams of kicking up a stink for sexual harassment because they consider it a normal form of human interaction. It’s also normal to compliment people on their looks and choice of attire. Both genders do it. Nobody gets offended. It’s more like “yeah, I put a ton of effort into choosing and matching this outfit this morning, and nobody noticed?! WTF??”

xellisx 8y ago

I thought the image was used to test formats well before JPEG.

Jun8 8y ago

In addition to lenna old timer image processing researchers would also have fond (depending whether your compression or edge detection algorithm worked correctly or not) memories of girl, cameraman, and barbara images as well as the infamous baboon (https://goo.gl/images/4Js7YH https://goo.gl/images/4Js7YH). I remember doing my senior thesis on 2D LPC in 1990 and since we did not know how to print images, in our report we just printed the pixel values as numbers. Ahh, the good old times.

toolslive 8y ago
infamous mandrill, I think it is. Anyway, that picture in full colour has a definite wow-factor. https://www.researchgate.net/figure/Original-standard-test-image-of-Mandrill-also-known-as-Baboon_fig9_259521525 https://www.researchgate.net/figure/Original-standard-test-i...

jordigh 8y ago

I feel that whenever Lena is mentioned, we should also give a shoutout to the researchers who were fed up with her and decided to use Fabio instead. https://arxiv.org/abs/1202.6429 https://arxiv.org/abs/1202.6429 What do you guys (yes, guys) think about making Fabio the standard image instead?

qubex 8y ago
Well on her (the researcher), her (Lena), and him (Fabio). Well on all of them.
- brokenmachine 8y ago
  You mean "good on them"?
johndough 8y ago
The Fabio image is not very useful wrt image compression: - grayscale (lenna image is rgb) - very odd 271x257 resolution (lenna image is square 512x512 which is a power of two) - just hair as fine detail features (lenna image has several: hat, feather boa and hair) https://4.bp.blogspot.com/-jUHXZ0jW4Lc/T1t3W0ex2BI/AAAAAAAAFeY/7TY0cTty1og/s400/Fabio.jpg https://4.bp.blogspot.com/-jUHXZ0jW4Lc/T1t3W0ex2BI/AAAAAAAAF... http://www.betapleatedchic.com/uploads/3/0/4/1/30413754/4907429_orig.gif http://www.betapleatedchic.com/uploads/3/0/4/1/30413754/4907... An optimal image would have the following properties: - different textures at different - scale - brightness - frequency - easily recognizable features - color - power of two resolution - no cultural issues - no licensing issues (not a scan of a magazine, globally recognized license, i.e. not "public domain") The problem with faces is that on the one hand, humans are able to tell easily if a feature was destroyed by an image compression algorithm, but on the other hand it's personally identifiable information and therefore comes with additional legal restrictions in many jurisdictions.
craigsmansion 8y ago
> we should also give a shoutout No, we shouldn't. Them using a different image in their research is perfectly acceptable--when one spends hours looking at something rendering, one shouldn't have to look at something one objects to--but referring to the research as a social statement diminishes the scientific value by framing it as a publicity stunt. Also, apart from using research time/money for licensing: "lena" has become shorthand for "face-female-with-various-patterns" and is a historical accident. "Fabio" will always be Fabio, a commercial model; a less than optimal choice for a scientific paper.

59nadir 8y ago

Lena Forsén* You'd think, having written a whole article about her, you'd be able to spell her name right. It's incredible to me how relaxed people in English speaking countries, particularly the US, are with removing and changing letters as they see fit. I remember seeing someone on Reddit, for example, who had named their kid Soren, after Søren Kierkegaard, completely ignoring that it's not even the same name. Do people realize that these letters are different looking for a reason?

rococode 8y ago
Isn't that just how language works? When a name doesn't suit your language's writing system it's natural to adapt it in a way that's more accessible. English simply doesn't have an acute accent. It's why people in English speaking countries are named Michael (or one of the 50+ alternatives) instead of the original מִיכָאֵל, or why BBC reports on Chinese President Xi Jinping instead of Chinese President 习近平.
- jabberthemutt 8y ago
  It is also natural to be attracted to sweet pictures of the other sex. But this whole article pushes the mindset that this is not ok. So I would argue that ignorance of proper spellings (in the same alphabet! Your examples are misleading) marginalizes them.
  - int_19h 8y ago
    It's not the same alphabet. English alphabet is distinct from the Swedish alphabet, they just happen to use the same writing system.
    - jabberthemutt 8y ago
      Sorry, second language English here. I thought it was obvious what I meant. é is very close in appearance and sound to e compared to some Hebrew(?) or Chinese.
- 8y ago

loosetypes 8y ago

I wasn't aware of the compared cases from the article such as Suzanne Vega, "The Mother of the MP3", or Susan Bennett, the namesake for Siri. But I have heard that a 3d model of Thom Yorke (Radiohead)'s head is the de facto standard for 3d printer trials.

abricot 8y ago
So it isn't Benchy?

5555624 8y ago

> Yet the model herself (right, now 67) remained mostly a mystery—until now. Not exactly, there is even a web site about it -- http://www.lenna.org/ http://www.lenna.org/ -- and both she and the image have Wikipedia entries.