Planet Four Talk

The Scale of them photos

  • edoxnard by edoxnard

    Hi guys does anyone know what scale these photos are - are they all the same size. Just trying to put what I'm looking all into perspective

    Cheers Ed

    Posted

  • p.titchin by p.titchin

    Hi ed, yes this is a much fraught question, but being a simplistic so-and so, I tend to think of them as roughly 200 metres across, when initially looking at them. Only if I'm trying to judge very small objects do I worry more about being more accurate, Another simple guide I use is that a very small item that is not horribly pixilated when you enlarge is roughly the size of a normal modest coffee table. The exact figures are gettable, but I personally find this sort of thinking more useful when I look at an image. Hope it helps ~ Pete

    Posted

  • Kitharode by Kitharode moderator in response to p.titchin's comment.

    Hi Pete. The 'fraughtness' continues! 😦

    I've got the images as nearer to 400m across. There are exceptions to the rule where 200m would be nearer the truth, but I think they are fairly rare. (All to do with 1x1 or 2x2 binning, whatever that is!). In the past when we've asked about scale, the problem was that the answer always involved pixels and we didn't know what pixels were worth. So I badgered the science team and eventually got the answer I was looking for.

    In the main (I was told) images are 840 x 648 pixels with 1 pixel = 49.6cm. Which gives 416.6m x 321.4m per image.

    We could compromise and say that images are 'about 300m across give or take a 100'. 😃

    Hi Ed. Nice to see you here. Whatever the truth of the matter, I hope what we've said so far helps give you something of the perspective you were looking for.

    Posted

  • p.titchin by p.titchin in response to Kitharode's comment.

    Thanks Kith, your third paragraph is sort of the answer I arrived at. In my mind I just think 'simples, it 's 200 to 400 metres across, my little boulders are about a metre or so across, unless I need to be more accurate, I stay off pixel measuring. Ed- now you can get your micrometer out and 'go to town! 😃 ~ "Slightly vague and simple" ~ Pete!

    Posted

  • edoxnard by edoxnard in response to p.titchin's comment.

    Hi thanks for your reply pete so each square we are doing about 200 meters of land on mars - that makes you wander how many squares of land there are that size on that planet . must admit some of the ones that are really dark do look like they will be pretty scary places to be well i would find them scary hehe

    Posted

  • Kitharode by Kitharode moderator

    If you fancy doing a bit of simple calculation to find the size of fans, boulders, spiders, or whatever, then you might want to have a look at this: http://talk.planetfour.org/#/boards/BPF0000008/discussions/DPF0000jfi

    Simples 😃

    Posted

  • Kitharode by Kitharode moderator in response to edoxnard's comment.

    Your question about how many squares of land (the size of a P4 image) there are on Mars tempted me to dust of my calculator and have a bit of fun.

    If we take the P4 images as 400(ish) x 300(ish) metres, then there are 7.5 images per square kilometre (just about).

    Surface area of Mars = 144,800,000 square km. Multiply by 7.5 gives us 1,086,000,000 images for the whole planet.

    It's a good job we're not trying to do that many. If 100 people each looked at 100 images every day, they would get through 3,650,000 images per year. Which means it would take 39.7 years to view the whole planet !! ** 😄 **

    Posted

  • DZM by DZM admin in response to Kitharode's comment.

    If you hadn't done that calculation, I totally would have! Thanks!! Pretty cool that it's just over 1 billion.

    Well, we have 4,461,316 classifications to date, although obviously the number of actually classified images is some small fraction of that...

    But still! We're making a dent! 😄

    Posted

  • HMB6EQUJ5 by HMB6EQUJ5 in response to DZM's comment.

    DZM. kindly clarify for me distinction between "classifications" and "actually classified". many thanks 😃

    Posted

  • p.titchin by p.titchin in response to HMB6EQUJ5's comment.

    Refer, igitur ego sum 😃 'Molis Petrus' 😃

    Posted

  • Kitharode by Kitharode moderator

    ... et ego homo austerus sum locus ?

    Bill. Each image needs to be classified by numerous people before it's 'retired' for analysis, so the 'actually classified' number is always smaller than the total number of classifications.

    Posted

  • HMB6EQUJ5 by HMB6EQUJ5 in response to Kitharode's comment.

    great! 😃 thanks

    Posted

  • mschwamb by mschwamb scientist, translator in response to Kitharode's comment.

    I thought y'all might want the actual number. Currently 30 people have to classify a Planet Four cutout before it is retired from rotation in the classification interface. That threshold number used to be higher earlier on for the first year of the project.

    Cheers,

    ~Meg

    Posted

  • mschwamb by mschwamb scientist, translator in response to Kitharode's comment.

    Kith's values are the ones that I get for the width and height of the 2x2 binning. Although I apparently accidentally added an extra zero or two in the P4 paper draft (so I better go fix that). A good chunk of the images are 49.6 cm/pixel resolution. The cutouts are 648 pixels in height and 840 pixels in width. Typically using the View HiRISE Image link there will be one number for the 'Original image scale range' and can then calculate the size of the image. This cutout looks to come from a HiRISE image where they used two binning modes on different parts of the image (not sure why there is two values in the image scale range. Usually it's only one number quoted.

    Cheers,

    ~Meg

    Posted

  • mschwamb by mschwamb scientist, translator in response to Kitharode's comment.

    Hi Kith,

    Funny thing is that recently the HIRISE team talked about this very subject. The HiRISE team wrote on their facebook page a similar calculation. They say it would take 350 years! Note: I haven't done the calculation myself. Details are on the P4 facebook page (I reposted the HiRISE post) if you want to double check the calculation.

    Cheers,

    ~Meg

    Posted

  • pete-j by pete-j

    Hi Meg

    I noticed the dual binning modes on the HIRISE images a few months ago and raised this point in the post I started in:
    Scaling Factors November 13th 2014.

    You may have noticed that sometimes out of interest I work out the #blotch or #fan size in the images. I have always done this when one binning mode is provided: either 1:1 or 2:2. Therefore avoiding the dual-binning modes, as I have just not had time to get round to reading this up more in-depth.

    Cheers,
    Pete

    Posted

  • mschwamb by mschwamb scientist, translator in response to pete-j's comment.

    Hi Pete,

    Cool that you're doing that. This is definitely something we want to do for the first paper even if it's just on average basis per region when we combine everyone's markings to make the fan and blotch catalog.

    We had a discussion of the binning on our science team telecon last week. Most of the images are 1x1. A smaller subset are 2x2.

    Cheers,

    ~Meg

    Posted

  • pete-j by pete-j

    Hi Meg,

    It would be interesting to compare the statistical distribution of blotch and fan sizes between the northern and southern hemispheres. The southern hemisphere would have the larger size (more extreme climate). Presumably this would mean a repeat experiment of the north and the Vatistas Borealis region?

    Cheers,

    Pete

    ps Nothing like a swim in the Oceanus Borealis... millions of years too late but couldn't resist putting this down!

    Posted

  • mschwamb by mschwamb scientist, translator in response to pete-j's comment.

    Hi Pete,

    There is a plan to head to the North. There are fans now spotted on dunes. The qualitative assessment is that they're smaller than the fans on the South Pole. Planet Four will be able to do a more quantitative comparison. A little more about the North from this blog post Anya wrote a while back and this post Anya wrote recently.

    Cheers,

    ~Meg

    Posted

  • wassock by wassock moderator in response to mschwamb's comment.

    "Explosions under the ice"???

    We have had a bit of a look at the Northern fans previously http://talk.planetfour.org/#/boards/BPF0000006/discussions/DPF0000e9m

    Posted

  • pete-j by pete-j

    Fan-tastic! Thanks for the links.

    I admit that I tend to spend more time classifying then reading around (when I have some spare time); an error on my part, although more recently I have strived to read through the site more before classifying...

    Meg - Apologies for going back to this, the calculations are simple to work out the image scale; but I still find the dual binning modes confusing particularly when both the 2x2 and 1x1 are included on the same HIRISE image; UNLESS the first binning mode is the primary one (and the one to use??) and the second one means that the image is at a different resolution?

    Posted

  • mschwamb by mschwamb scientist, translator in response to pete-j's comment.

    No worries. That's part of my role to help describe the science and point to resources we have. I'm happy to try and answers these questions, so keep them coming Our great Talk moderators are also great resources as well as some of our veterans.

    As for the two binning modes i don't actually know what it specifically means. I'll see if someone else on the science team knows.

    Cheers,

    ~Meg

    Posted

  • Portyankina by Portyankina scientist

    Hi!
    Each HiRISE image is made of several channels (2 channels for each of IR and BG and each of RE strips). Each channel is composed of 2 CCDs. Sometimes the team goes overly complex and does different binning for each of those channels or even CCDs. This is what you most probably have spotted. One has to figure out what is the sequence in which the binning of the CCDs are mentioned in the label file...

    Anya

    Posted

  • pete-j by pete-j

    Hi Anya

    Thank you for your response. I take it that IR stands for Infra-Red Continuum and BG Blue Green Continuum?

    Basically, you mean either (1x1) or (2x2) may be the primary binning mode on such an image?

    In this respect, I generally use an image with a single binning mode (1x1) or (2x2) and then try to work out the size of the features in the images and avoid the dual binning modes. I don't handle such images on a daily basis and be read-up on the details. Thanks...

    Cheers,

    Pete

    Posted