Jump to content

Tree counting


qed

Recommended Posts

I getting some images of a oil palm estate taken from a DIY drone. The image is stitch and mosaic. Has anyone done automated tree counting on these images.

 

Seen some examples using eCognition but that was with multi-spectral images.

 

  • Thanks 1
Link to comment
Share on other sites

If you have a very-high resolution image (< 1m ) and it has only 3 spectral bands in the visible spectrum (Red, Green, Blue - RGB) with no additional infrared bands (NIR), the best option to extract trees is to use an OBIA method ( > eCognition is one of the best OBIA softwares).

 

So, for your task, me I would use eCognition.

 

;)

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

  • 3 weeks later...

look at maxmax website, they do NIR conventions on cameras that can fit on drones. Did a canon compact with them 100%. Try imageJ (free software), run the NDVI settings you will get some data. Or use a blue filter on your camera, you will get usable images for NDVI. 

I got good data with dessert palms from RGB images in eCognition, but you have o play with the analysis a bit! Having a NIR layer will make life easy for you. 

 

Good Luck ;)

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

  • 4 weeks later...

I getting some images of a oil palm estate taken from a DIY drone. The image is stitch and mosaic. Has anyone done automated tree counting on these images.

 

Seen some examples using eCognition but that was with multi-spectral images.

Does the tree comes with a different color than all the other object? If yes then try supervised classification creating two class one with the color of the tree and other for all image. it will classify the image distinct color to the tree. then count the occurrence of the color. But it is not a good process. from image it may not be possible to count the tree until you have the signature of tree. If you have NIR (Near Infra Red) image you will get a patch of tree not a count.  

Link to comment
Share on other sites

Oil palm tree counting is a quite simple application. You can simply use following methods:

 

1. Ravine extraction (Honda Kyioshi): Based on the fact that the top canopy is the brightest point. The ravine (or the valley between 2 trees) looks darker. Doing it, you wil be able to locate the central point of each tree. Then counting is some simple works to do aftewards.

2. Local threshold texture matching: Requires some programming tasks. Get the individual trees as samplings (with different sizes of canopy) and then match them over the whole image. The matching algorithm can be image correlation or histograming matching. I already did it with VB programming and it works perfectly. By doing this, you can also be able to map individual trees.

 

All of this is based on RGB image. For better results, mask the plantation areas before running the procedure.

 

Of course there are several other options to do. If you want to operate the procedure in an automatic way, the programming skill is needed.

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

  • 1 month later...

It might not be a very good idea to use ArcGIS but somehow we can use this software for tree counting!

 

My method is quite simple:

 

1. Get a piece of hi-res image from Google Earth

 

lucngan.jpg

 

 

2. Open the image in ArcGIS

3. Make sure toolbar "Image Classification" is checked (turned on).

4. From "Image Classification" toolbar, select "Classification" --> "Iso Cluster Unsupervised Classification" command.

5. Chose "2" for "Number of Classes" in the dialogbox. By doing it, you will have only trees (or something else similar) and non-tree objects in the result. Run it and see the result (below).

 

exp1.jpg

 

C'mon, it also shows "Google Earth" trademark on the result (LOL).

6. Now go to ArcGIS ToolBox, select "Conversion" --> "From Raster" --> "Raster to Polygon". In the dialog box, check to "Simplify Polygion (optional)" checkbox.

7. The result is as below

 

out2.jpg

 

8. Mask the area where you want to count the trees, then count the number of polygons within that mask. Bingo!!!!

 

If you want to do it better, you can do some pre-processing steps for your image. Also, when you have the polygon layer, you can try to simplify the layer again (using Eliminate, Integrate... functions) before counting trees.

 

Enjoy ESRI.

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

  • 3 months later...
  • 1 month later...
  • 1 month later...

look at maxmax website, they do NIR conventions on cameras that can fit on drones. Did a canon compact with them 100%. Try imageJ (free software), run the NDVI settings you will get some data. Or use a blue filter on your camera, you will get usable images for NDVI. 

I got good data with dessert palms from RGB images in eCognition, but you have o play with the analysis a bit! Having a NIR layer will make life easy for you. 

 

Good Luck ;)

Hi Green_wing, I have a converted Canon SX260 (from maxmax) with NIR,G,B filters. I mosaic my images (from drone, 50m. AGL.)  then generate NDVI's with Imagej. How do I produce the Tree count in eCognition ?. Thx. Dave

Link to comment
Share on other sites

  • 2 weeks later...
  • 3 months later...

just share my experience

 

once I ever did the same task for palm counting from quickbird imagery.

I only use the red band (green also can) and did some supervised classification with proper tree sample of course.

after that, simply convert the result to shapefile and did some conversion to centroid point.

my result is about at 94% accuracy.

but it only easy to run on palm with not very dense canopy.

 

good luck

Link to comment
Share on other sites

  • 1 year later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.

Disable-Adblock.png

 

If you enjoy our contents, support us by Disable ads Blocker or add GIS-area to your ads blocker whitelist