Auto tree counting for oil palm plantation based on UAV/drone orthophoto /
Eiffel Syafhamka Nadimen A/l Syamsul,
Auto tree counting for oil palm plantation based on UAV/drone orthophoto / Eiffel Syafhamka Nadimen A/l Syamsul - xii, 61 pages : illustrations ; 1 CD-ROM.
Center for Mathematical Sciences
Bachelor of Applied Science In Data Analytics With Honour -- Universiti Malaysia Pahang – 2023
Includes bibliographical references
Tree counting plays a crucial role in assessing the number of tree crops within a specific area or plantation farms, and computer vision techniques using aerial or satellite imagery have become prevalent in this domain. Currently, manual digitization is the traditional method employed for tree counting, but it falls behind the advancements in machine learning. To address this, a more effective approach is sought to assist plantation owners with palm oil tree counting and land resource management for a faster and more accurate tree counting method that can be used in ArcGIS. This study will utilize the Faster R-CNN due to its usage of a two-stage pipeline and leverage power of deep learning. This method was able to detect and plot the palm oil tree on the crown of each tree but due to the average precision only being 75%, the method can only detect the trees correctly to a certain extent. Also, the transformation of georeferencing plot points for all the tree using an affine transformation coefficient causes the output to not be aligned in its correct position when outputted into ArcGIS. With that the deployment of Faster RCNN method for palm oil tree counting has been successful, demonstrating its ability to offer a faster and more efficient alternative to manual digitize but the integration of output into ArcGIS software poses as a challenge. The result of implementing this method may be helpful in cutting the cost of labour, saving the time spent to manually digitize the tree counting method and improving the productivity of work.
THE0010045 (Local)
Center for Mathematical Sciences--Dissertations
Universities and colleges--Dissertations
Final Year Report--Dissertations
Auto tree counting for oil palm plantation based on UAV/drone orthophoto / Eiffel Syafhamka Nadimen A/l Syamsul - xii, 61 pages : illustrations ; 1 CD-ROM.
Center for Mathematical Sciences
Bachelor of Applied Science In Data Analytics With Honour -- Universiti Malaysia Pahang – 2023
Includes bibliographical references
Tree counting plays a crucial role in assessing the number of tree crops within a specific area or plantation farms, and computer vision techniques using aerial or satellite imagery have become prevalent in this domain. Currently, manual digitization is the traditional method employed for tree counting, but it falls behind the advancements in machine learning. To address this, a more effective approach is sought to assist plantation owners with palm oil tree counting and land resource management for a faster and more accurate tree counting method that can be used in ArcGIS. This study will utilize the Faster R-CNN due to its usage of a two-stage pipeline and leverage power of deep learning. This method was able to detect and plot the palm oil tree on the crown of each tree but due to the average precision only being 75%, the method can only detect the trees correctly to a certain extent. Also, the transformation of georeferencing plot points for all the tree using an affine transformation coefficient causes the output to not be aligned in its correct position when outputted into ArcGIS. With that the deployment of Faster RCNN method for palm oil tree counting has been successful, demonstrating its ability to offer a faster and more efficient alternative to manual digitize but the integration of output into ArcGIS software poses as a challenge. The result of implementing this method may be helpful in cutting the cost of labour, saving the time spent to manually digitize the tree counting method and improving the productivity of work.
THE0010045 (Local)
Center for Mathematical Sciences--Dissertations
Universities and colleges--Dissertations
Final Year Report--Dissertations