Loading ...
Sorry, an error occurred while loading the content.
 

Stitching a 32646 x 27955 image

Expand Messages
  • rox_ana_mustata
    In an experimental work I ve made a very large panorama composed from 230 photos. The PTGui software was very good in handling all the images in the stitching
    Message 1 of 6 , Jul 31 10:28 AM
      In an experimental work I've made a very large panorama composed from
      230 photos. The PTGui software was very good in handling all the images
      in the stitching process. Details regarding the workflow can be found
      in the webpage:

      http://www.panoramas.ro/tutorials/large-pano.html

      Comments and questions are welcome.
    • John Riley
      ... That is a fascinating experiment you have done. I assume that you were moving the sample about, making rows and columns of photos. This seems similar to
      Message 2 of 6 , Jul 31 8:00 PM
        On Jul 31, 2007, at 1:28 PM, rox_ana_mustata wrote:

        > In an experimental work I've made a very large panorama composed from
        > 230 photos. The PTGui software was very good in handling all the
        > images
        > in the stitching process. Details regarding the workflow can be found
        > in the webpage:
        >
        > http://www.panoramas.ro/tutorials/large-pano.html
        >
        > Comments and questions are welcome.


        That is a fascinating experiment you have done. I assume that you
        were moving the sample about, making rows and columns of photos.
        This seems similar to the type of work we have seen where someone
        photographs a street or a mural to create a "linear panorama", except
        you have the added complexity of multi-row photography! Folks who
        make linear panoramas, if I recall correctly, have to do a lot of
        work because of the shifting of the camera and the resulting parallax
        problems. At least here you don't have the degree of depth issues
        (like side streets) that the usual linear pano must deal with.

        If I am understanding properly what you did, it is remarkable that
        PTGui handled it so well on "automatic"! You also seem to have very
        good registration between the different color mosaics. That is
        surprising, since you shot each color separately it seems. How did
        you combine the colors to make the composite image? (Also, what kind
        of tissue was it?)

        Excellent, interesting work,

        John



        John C. Riley, Ph.D.
        Assoc. Prof. of Physics
        Univ. of SC Upstate

        and 4Pi-VR media solutions

        jriley@...
        johnriley@...
        864-431-7075 (c)
        864-503-5775 (w)
        864-461-3504 (h)







        [Non-text portions of this message have been removed]
      • rox_ana_mustata
        The sample was moved under the microscope objective. For each move, three photos were acquired (blue, red, and green color, corresponding to three fluorescence
        Message 3 of 6 , Aug 2, 2007
          The sample was moved under the microscope objective. For each move,
          three photos were acquired (blue, red, and green color, corresponding
          to three fluorescence filter sets). The image acquisition was a pain in
          the ass, taking 3 days (with an average of ~5 hours daily). As you can
          see, should be 231 images for each color (one image is missing, look in
          the middle left part, of the section).
          First I've tried with PTGui 6 and a lot of pictures were not correctly
          positioned. PTGui 7 is better regarding the control points system.
          The combined image was obtained by layering the images in PS. The upper
          layers were set to difference, resulting in combination of colors (for
          example, green + red makes yellow).
          The sample was a section from a human mole. The medical diagnostic is
          not relevant here. The coloring procedures done here are only
          experimental.

          Roxana
        • David Sykes
          ... I guess the only realistic option is 10 rows of 23 images each. The aspect-ratio of the final image implies a lot of horizontal overlap and hardly any
          Message 4 of 6 , Aug 3, 2007
            rox_ana_mustata wrote:
            >
            > In an experimental work I've made a very large panorama composed from
            > 230 photos.

            I guess the only realistic option is 10 rows of 23 images each.
            The aspect-ratio of the final image implies a lot of horizontal
            overlap and hardly any vertical overlap.

            Is that correct ?

            What is the NA of the x20 objective, the relay lens (or
            eyepiece/camera lens combination) magnification and pixel-size of the
            camera sensor ?


            > Comments and questions are welcome.


            There is a low-resolution web image but no link for the cropped
            high-resolution image.



            David Sykes
          • rox_ana_mustata
            Is not such simple… There were 22 rows, having from 1 image (row 1, the top) to 13 images. As far as I remember the objective NA is 0.75. The microscope is a
            Message 5 of 6 , Aug 4, 2007
              Is not such simple… There were 22 rows, having from 1 image (row 1,
              the top) to 13 images.
              As far as I remember the objective NA is 0.75. The microscope is a
              Nikon E600 with a DS-U1 digital camera. Take a look at
              http://www.nikonusa.com/template.php?cat=5&grp=26&productNr=DS5MU1
              for technical specification.
              I updated the webpage with a 100% crop image from the composite .jpg
              file (21408 x 25000 pixels).

              Roxana

              --- In PanoToolsNG@yahoogroups.com, "David Sykes" <killspammers2@...>
              wrote:

              > I guess the only realistic option is 10 rows of 23 images each.
              > The aspect-ratio of the final image implies a lot of horizontal
              > overlap and hardly any vertical overlap.
              >
              > Is that correct ?
              >
              > What is the NA of the x20 objective, the relay lens (or
              > eyepiece/camera lens combination) magnification and pixel-size of
              the
              > camera sensor ?
              >
              > There is a low-resolution web image but no link for the cropped
              > high-resolution image.
              >
              >
              >
              > David Sykes
              >
            • David Sykes
              ... I make that 286 images, not 230 :-) ... Thanks, but that is the camera control unit. Which camera head is it ? David Sykes
              Message 6 of 6 , Aug 5, 2007
                Roxana wrote:

                > There were 22 rows, having from 1 image ... to 13 images.

                I make that 286 images, not 230 :-)

                > The microscope is a Nikon E600 with a DS-U1 digital camera.


                Thanks, but that is the camera control unit.


                Which camera head is it ?


                David Sykes
              Your message has been successfully submitted and would be delivered to recipients shortly.