Hi LokeRichard suggested using a NDVI approach which would require data in the 
Near Infrared (NIR) band.  An orthomisaic would contain 3 colour bands, but not 
the NIR.  NDVI provides an understanding of plant greenness and health and may 
not be suitable for detecting oil palm's.  If you are getting 90 percent plus 
with your current approach , you may find that you will spend 4 times longer 
capturing the additional 10 percent.  NIR data,are available from various 
satellite sensors however, their resolution would not work at the individual 
plant level.Some drone sensors record the NIR band.Kirk Schmidt Sent from my 
Galaxy
-------- Original message --------From: Loke Jian Munn <[email protected]> 
Date: 2023-06-06  6:18 a.m.  (GMT-04:00) To: Richard McDonnell 
<[email protected]>, kirk <[email protected]>, 
[email protected] Cc: Moke Kwai Cheong <[email protected]>, Cheong 
Chee Phang <[email protected]>, Leong Kah Choon <[email protected]>, 
Mulwanto <[email protected]>, Endang Sudita Prayoga <[email protected]>, M 
Rizki Ramdani <[email protected]>, Fira Yunita <[email protected]>, Qurata Ayun 
Wijanarko <[email protected]> Subject: RE: [Qgis-user] Palm Counting using 
QGIS 

Hi Kirk & Richard,
 
Thanks for the great inputs and we have a clearer direction now.
 
Just to have a quick check with you, can the Imagery be Orthomosaic map ?
 
Thank you.
 
Best Regards,

Loke Jian Munn
Information Technology (IT) Department

KUALA LUMPUR KEPONG BERHAD
Wisma Taiko,
No. 1, Jalan S.P. Seenivasagam,
30000 Ipoh,Perak Darul Ridzuan,Malaysia
T    : 
+605 240 8000 (ext. 2443)


 


From: Richard McDonnell <[email protected]>

Sent: Thursday, 1 June, 2023 10:55 PM
To: kirk <[email protected]>; Loke Jian Munn <[email protected]>; 
[email protected]
Cc: Moke Kwai Cheong <[email protected]>; Cheong Chee Phang 
<[email protected]>; Leong Kah Choon <[email protected]>; Mulwanto 
<[email protected]>; Endang Sudita Prayoga <[email protected]>; M Rizki 
Ramdani <[email protected]>; Fira Yunita <[email protected]>;
 Qurata Ayun Wijanarko <[email protected]>
Subject: RE: [Qgis-user] Palm Counting using QGIS


 
Hi Loke,
Just to add to Kirk’s comments, have you considered using Imagery along with 
the DSM? More specifically a NDVI of the
 plantation.
You will need to determine the signature range for Palm Tree Leaves, but where 
you have both a Palm tree Signature and
 your Local Minima/Maxima it could improve your detection.
Kind Regards,
 
Richard

 





——
Richard McDonnell MSc GIS, FME Certified Professional
FRM Data Management

——
Oifig na nOibreacha Poiblí
Office of Public Works

Sráid Jonathan Swift, Baile Átha Troim, Co na Mí, C15 NX36

Jonathan Swift Street, Trim, Co Meath, C15 NX36
——
M +353 87 688 5964 T +353 46 942 2409
https://gov.ie/opw

——
To send me files larger than 30MB, please use the link below 
https://clicktime.symantec.com/15uBY3eXTuVgDRHFkBtiB?h=f_sNL5U9oSfDKaodrCLg96KXSTHDaiQO0T0VIEDipgY=&u=https://filetransfer.opw.ie/filedrop/richard.mcdonnell%40opw.ie

Email Disclaimer: 
https://clicktime.symantec.com/15uBhi36P8rs3Jw6qJh1R?h=7BRo4J_IELubeqOZ5acMrrBwdwGb5BtTEWtxK9bd_8w=&u=https://www.gov.ie/en/organisation-information/439daf-email-disclaimer/





 





——
MSc GIS, FME Certified Professional

——
Oifig na nOibreacha Poiblí
Office of Public Works

Sráid Jonathan Swift, Baile Átha Troim, Co na Mí, C15 NX36

Jonathan Swift Street, Trim, Co Meath, C15 NX36
——
M +353 87 688 5964 T +353 46 942 2409
https://https://gov.ie/opw

——
Email Disclaimer: 
https://www.gov.ie/en/organisation-information/439daf-email-disclaimer/ 








From: QGIS-User <[email protected]>
On Behalf Of kirk via QGIS-User
Sent: 01 June 2023 10:58
To: Loke Jian Munn <[email protected]>;
[email protected]
Cc: Moke Kwai Cheong <[email protected]>; Cheong Chee Phang 
<[email protected]>; Leong Kah Choon <[email protected]>; Mulwanto
 <[email protected]>; Endang Sudita Prayoga <[email protected]>; M Rizki 
Ramdani <[email protected]>; Fira Yunita <[email protected]>;
 Qurata Ayun Wijanarko <[email protected]>
Subject: Re: [Qgis-user] Palm Counting using QGIS


 

Hi Loke.


 


A couple of comments.  


 


1) The 95 % success rate is very good.  If you were to have a close look at the 
missed trees you may realize that the highest point on a particular palm tree 
is lower than the neighboring tree or palm frond. What ever
 algorithm you use, you are limited by the physical data, accuracy and 
resolution being the key factors.  The physical condition of the trees you are 
trying to identify also plays a role,as mentioned previously.  A DSM is an 
approximation as it is derived from
 some form of point data, either lidar  or photo based and then generated by 
extracting the maximum, 95th percentile, maximum or perhaps the average height 
if all the points that land in a cell.  


 


Your issue is a limitation of the data and the basic logic of a max/minima 
algorithm.  It cannot detect what is not reflected in the data.  You can 
interpret the colour imagery but have a look at the dsm at the missed
 maxima.


 


You could try developing a routine by processing the,raw point cloud data or 
try generating a dsm yourself.  


 


You can also try changing the algorithm parameters. Perhaps in areas if missed 
maxima. Although, this may not be effective as the maxima you are looking for 
probably is not reflected the dem.


 


 


 


2) Plug-ins,are created by third parties and not by the Qgis team directly. You 
can reach out to the author and discuss with him/her. 


 


Kirk Schmidt


 


 


 


 



Sent from my Galaxy



 


 


-------- Original message --------


From: Loke Jian Munn via QGIS-User <[email protected]>



Date: 2023-06-01 6:28 a.m. (GMT-04:00)



To: [email protected]



Cc: Moke Kwai Cheong <[email protected]>, Cheong Chee Phang 
<[email protected]>,
 Leong Kah Choon <[email protected]>, Mulwanto <[email protected]>, Endang 
Sudita Prayoga <[email protected]>,
 M Rizki Ramdani <[email protected]>, Fira Yunita <[email protected]>, Qurata 
Ayun Wijanarko <[email protected]>



Subject: [Qgis-user] Palm Counting using QGIS



 


Hi Team,
 
We have been using QGIS to do Palm Counting.

 
Palm Counting here mean oil palm, we use QGIS processing tool, Features-Raster 
Tools (Local Minima and Maxima) to do the counting on a DSM layer.
Example as below:
 

 
Based on our industrial experiences, the accuracy of the Local Maxima algorithm 
is approximately 90%-95%. However, we still need to do toggle editing such as
 we utilize the “ Add Point Feature “ function when there are points missed, 
the “Vertex Tool” is used to adjust the position of points and the “Select 
Feature” for removal of any unnecessary points.
 
We would like to have an improvement plan on the accuracy for
more than 95% to reduce the extra manual works ( toggle edit ).
 
Appreciate if the team could share with us any latest QGIS palm count plugin or 
any other solutions to achieve the accuracy more than 95% when doing palm count.
 
Your input is highly appreciated. Thank you.
 
 
Regards,
Loke Jian Munn
Information Technology (IT) Department

KUALA LUMPUR KEPONG BERHAD

Wisma Taiko,

No. 1, Jalan S.P. Seenivasagam,

30000 Ipoh,Perak Darul Ridzuan,Malaysia

T    : 
+605 240 8000 (ext. 2443)


 

DISCLAIMER : This email and any attachments are confidential and intended 
solely for the addressee. If you are not the intended recipient, any 
disclosure, copying, distribution either whole or partial, is prohibited and 
may be unlawful. In such case, please
 delete this message and kindly notify the sender. 
Kuala Lumpur Kepong Berhad 
 
 
 


DISCLAIMER : This email and any attachments are confidential and intended 
solely for the addressee. If you are not the intended recipient, any 
disclosure, copying, distribution either whole or partial, is prohibited and 
may be unlawful. In such case, please
 delete this message and kindly notify the sender. 
Kuala Lumpur Kepong Berhad 



DISCLAIMER : This email and any attachments are confidential and intended 
solely for the addressee. If you are not the intended recipient, any 
disclosure, copying, distribution either whole or partial, is prohibited and 
may be unlawful. In such case, please
 delete this message and kindly notify the sender. 
Kuala Lumpur Kepong Berhad 


_______________________________________________
QGIS-User mailing list
[email protected]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Reply via email to