Dear ImageJ users:
A colleague of mine and I are calculating areas of fossil leaves. These fossils often do not have a lot of contrast with the rest of the solid (the so-called "rock" or, more technically,matrix). I have two questions: a. Is there a way to enhance the contrast of the leaf with respect to the matrix so that we can trace it with more confidence? If "yes", what is the sequence of commands in ImageJ. b. Sometimes the fossil is located on the border of the "rock". Is there a way to tell ImageJ to automatically follow the edge of the rock until it finds the origin of the area trace? It would be faster and more exact than us, humans, doing it by hand. Gratefully, Jorge Jorge A. Santiago-Blay, PhD *https://blaypublishers.com <http://blaypublishers.com>* 1. Positive experiences for authors of papers published in *LEB* http://blaypublishers.com/testimonials/ 2. Free examples of papers published in *LEB*: http://blaypublishers.com/category/previous-issues/. 3. *Guidelines for Authors* and page charges of *LEB*: http://blaypublishers.com/archives/ *.* 4. Want to subscribe to *LEB*? http://blaypublishers.com/subscriptions/ http://blayjorge.wordpress.com/ http://paleobiology.si.edu/staff/individuals/santiagoblay.cfm -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Greetings Jorge,
without seeing *representative* (not best) samples of your data, it is impossible to give *reliable* advice. Perhaps the most important step in image analysis is image acquisition. How do you go about (illumination, colour, camera, optics)? Did you try to enhance the contrast by using oblique illumination or by coloured light? In any case, please optimize your image acquisition process by using professional equipment and techniques, because post hoc image processing in most cases can't compensate for suboptimum image acquisition and if, it will become rather costly. Regards Herbie :::::::::::::::::::::::::::::::::::::::::::::::::::: Am 15.10.20 um 15:24 schrieb Jorge A. Santiago-Blay: > Dear ImageJ users: > > A colleague of mine and I are calculating areas of fossil leaves. These > fossils often do not have a lot of contrast with the rest of the solid (the > so-called "rock" or, more technically,matrix). I have two questions: > > a. Is there a way to enhance the contrast of the leaf with respect to the > matrix so that we can trace it with more confidence? If "yes", what is the > sequence of commands in ImageJ. > > b. Sometimes the fossil is located on the border of the "rock". Is there a > way to tell ImageJ to automatically follow the edge of the rock until it > finds the origin of the area trace? It would be faster and more exact than > us, humans, doing it by hand. > > Gratefully, > > Jorge > > Jorge A. Santiago-Blay, PhD > *https://blaypublishers.com <http://blaypublishers.com>* > > 1. Positive experiences for authors of papers published in *LEB* > http://blaypublishers.com/testimonials/ > > 2. Free examples of papers published in *LEB*: > http://blaypublishers.com/category/previous-issues/. > > 3. *Guidelines for Authors* and page charges of *LEB*: > http://blaypublishers.com/archives/ *.* > > 4. Want to subscribe to *LEB*? http://blaypublishers.com/subscriptions/ > > http://blayjorge.wordpress.com/ > > http://paleobiology.si.edu/staff/individuals/santiagoblay.cfm > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Jorge,
without going into detailed investigations, here is my basic question: How do you expect to get area measures if the leaf fossils are incomplete and, as in the provided samples, two of them partly overlap. In any case, it will be near to impossible to automatically detect the (broken) edges of the leaf fossils in the sample image (especially at the top left of the top layer). Please consider altering your image acquisition process (structured / coloured light). I try to attach an achromatic selection of the original sample image (which was a JPG-compressed colour image without much colour information). Please note that in general JPG-compressed images are not suited for scientific purposes. Furthermore, the fossil-parts appear being not perfectly focused. Which camera and camera-optics do you use? Perhaps others on the list can provide further help. Regards Herbie :::::::::::::::::::::::::::::::::::::::::::::::::::: Am 15.10.20 um 16:48 schrieb Jorge A. Santiago-Blay: > > > ---------- Forwarded message --------- > From: *Jorge A. Santiago-Blay* <[hidden email] > <mailto:[hidden email]>> > Date: Thu, Oct 15, 2020 at 10:46 AM > Subject: Fwd: Borders of fossil leaves: enhancing their margins and edges > To: <[hidden email] <mailto:[hidden email]>> > > > Hi Herbie: > > Thanks for your feedback. Below are examples of what I meant: > > F-705 Scale_Jorge.jpg > . > Note, I do not know if ImageJ likes attachment and the requirements, > hence my imageless email. > > If you wish to reply, my email is [hidden email] > <mailto:[hidden email]> > > Gratefully, > > Jorge > > ---------- Forwarded message --------- > From: *Herbie* <[hidden email] <mailto:[hidden email]>> > Date: Thu, Oct 15, 2020 at 10:12 AM > Subject: Re: Borders of fossil leaves: enhancing their margins and edges > To: <[hidden email] <mailto:[hidden email]>> > > > Greetings Jorge, > > without seeing *representative* (not best) samples of your data, it is > impossible to give *reliable* advice. > > Perhaps the most important step in image analysis is image acquisition. > How do you go about (illumination, colour, camera, optics)? > > Did you try to enhance the contrast by using oblique illumination or by > coloured light? > > In any case, please optimize your image acquisition process by using > professional equipment and techniques, because post hoc image processing > in most cases can't compensate for suboptimum image acquisition and if, > it will become rather costly. > > Regards > > Herbie > > :::::::::::::::::::::::::::::::::::::::::::::::::::: > Am 15.10.20 um 15:24 schrieb Jorge A. Santiago-Blay: > > Dear ImageJ users: > > > > A colleague of mine and I are calculating areas of fossil leaves. These > > fossils often do not have a lot of contrast with the rest of the > solid (the > > so-called "rock" or, more technically,matrix). I have two questions: > > > > a. Is there a way to enhance the contrast of the leaf with respect > to the > > matrix so that we can trace it with more confidence? If "yes", what > is the > > sequence of commands in ImageJ. > > > > b. Sometimes the fossil is located on the border of the "rock". Is > there a > > way to tell ImageJ to automatically follow the edge of the rock until it > > finds the origin of the area trace? It would be faster and more exact > than > > us, humans, doing it by hand. > > > > Gratefully, > > > > Jorge > > > > Jorge A. Santiago-Blay, PhD > > *https://blaypublishers.com <http://blaypublishers.com>* > > > > 1. Positive experiences for authors of papers published in *LEB* > > http://blaypublishers.com/testimonials/ > > > > 2. Free examples of papers published in *LEB*: > > http://blaypublishers.com/category/previous-issues/. > > > > 3. *Guidelines for Authors* and page charges of *LEB*: > > http://blaypublishers.com/archives/ *.* > > > > 4. Want to subscribe to *LEB*? http://blaypublishers.com/subscriptions/ > > > > http://blayjorge.wordpress.com/ > > > > http://paleobiology.si.edu/staff/individuals/santiagoblay.cfm > > > > -- > > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > > -- > Jorge A. Santiago-Blay, PhD > _https://blaypublishers.com <http://blaypublishers.com>_ > > 1. Positive experiences for authors of papers published in */LEB/* > http://blaypublishers.com/testimonials/ > > 2. Free examples of papers published in */LEB/*: > http://blaypublishers.com/category/previous-issues/. > > 3. /Guidelines for Authors/ and page charges of */LEB/*: > http://blaypublishers.com/archives//./ > > 4. Want to subscribe to */LEB/*? http://blaypublishers.com/subscriptions/ > > http://blayjorge.wordpress.com/ > > http://paleobiology.si.edu/staff/individuals/santiagoblay.cfm > > > -- > Jorge A. Santiago-Blay, PhD > _https://blaypublishers.com <http://blaypublishers.com>_ > > 1. Positive experiences for authors of papers published in */LEB/* > http://blaypublishers.com/testimonials/ > > 2. Free examples of papers published in */LEB/*: > http://blaypublishers.com/category/previous-issues/. > > 3. /Guidelines for Authors/ and page charges of */LEB/*: > http://blaypublishers.com/archives//./ > > 4. Want to subscribe to */LEB/*? http://blaypublishers.com/subscriptions/ > > http://blayjorge.wordpress.com/ > > http://paleobiology.si.edu/staff/individuals/santiagoblay.cfm ImageJ mailing list: http://imagej.nih.gov/ij/list.html F-705 Scale_Jorge.png (2M) Download Attachment |
I agree with Herbie that your time is better spent trying to improve the imaging than doing post-processing.
What do you get if you take an infrared image of the rocks? Perhaps that will increase the contrast? Or vice versa, if the leaves don't absorb infrared, you can subtract the infrared image from the normal image and be left with an image of only the leaves which will be easier to process. Stein -----Original Message----- Sent: 15. oktober 2020 17:16 Subject: Re: Fwd: Borders of fossil leaves: enhancing their margins and edges Jorge, without going into detailed investigations, here is my basic question: How do you expect to get area measures if the leaf fossils are incomplete and, as in the provided samples, two of them partly overlap. In any case, it will be near to impossible to automatically detect the (broken) edges of the leaf fossils in the sample image (especially at the top left of the top layer). Please consider altering your image acquisition process (structured / coloured light). I try to attach an achromatic selection of the original sample image (which was a JPG-compressed colour image without much colour information). Please note that in general JPG-compressed images are not suited for scientific purposes. Furthermore, the fossil-parts appear being not perfectly focused. Which camera and camera-optics do you use? Perhaps others on the list can provide further help. Regards Herbie :::::::::::::::::::::::::::::::::::::::::::::::::::: Am 15.10.20 um 16:48 schrieb Jorge A. Santiago-Blay: > > > ---------- Forwarded message --------- > From: *Jorge A. Santiago-Blay* <[hidden email] > <mailto:[hidden email]>> > Date: Thu, Oct 15, 2020 at 10:46 AM > Subject: Fwd: Borders of fossil leaves: enhancing their margins and > edges > To: <[hidden email] <mailto:[hidden email]>> > > > Hi Herbie: > > Thanks for your feedback. Below are examples of what I meant: > > F-705 Scale_Jorge.jpg > . > Note, I do not know if ImageJ likes attachment and the requirements, > hence my imageless email. > > If you wish to reply, my email is [hidden email] > <mailto:[hidden email]> > > Gratefully, > > Jorge > > ---------- Forwarded message --------- > From: *Herbie* <[hidden email] <mailto:[hidden email]>> > Date: Thu, Oct 15, 2020 at 10:12 AM > Subject: Re: Borders of fossil leaves: enhancing their margins and > edges > To: <[hidden email] <mailto:[hidden email]>> > > > Greetings Jorge, > > without seeing *representative* (not best) samples of your data, it is > impossible to give *reliable* advice. > > Perhaps the most important step in image analysis is image acquisition. > How do you go about (illumination, colour, camera, optics)? > > Did you try to enhance the contrast by using oblique illumination or > by coloured light? > > In any case, please optimize your image acquisition process by using > professional equipment and techniques, because post hoc image > processing in most cases can't compensate for suboptimum image > acquisition and if, it will become rather costly. > > Regards > > Herbie > > :::::::::::::::::::::::::::::::::::::::::::::::::::: > Am 15.10.20 um 15:24 schrieb Jorge A. Santiago-Blay: > > Dear ImageJ users: > > > > A colleague of mine and I are calculating areas of fossil leaves. > These > fossils often do not have a lot of contrast with the rest of > the solid (the > so-called "rock" or, more technically,matrix). I > have two questions: > > > > a. Is there a way to enhance the contrast of the leaf with respect > to the > matrix so that we can trace it with more confidence? If > "yes", what is the > sequence of commands in ImageJ. > > > > b. Sometimes the fossil is located on the border of the "rock". Is > there a > way to tell ImageJ to automatically follow the edge of the > rock until it > finds the origin of the area trace? It would be > faster and more exact than > us, humans, doing it by hand. > > > > Gratefully, > > > > Jorge > > > > Jorge A. Santiago-Blay, PhD > > -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Free forum by Nabble | Edit this page |