Document Type : Research Article


1 Assistant Professor, Department of Biology, College of Natural and Computational Science, DambiDollo University, DambiDollo, Oromia Region, Ethiopia.

2 Physics Department, DambiDollo University, DambiDollo, Oromia Region, Ethiopia.

3 Assistant Professor, Department of Computer Science, College of Engineering and Technology, DambiDollo University, DambiDollo, Oromia Region, Ethiopia

4 Professor, Department of CSE, Sri Eshwar College of Engineering (Autonomous), Coimbatore, Tamil Nadu, India.

5 Associate Professor, Department of EEE, Sona College of Technology, Salem, Tamil Nadu, India.

6 Associate Professor, Department of Analytics , School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, Tamil Nadu, India.


Massive falling of leaves in dense tree region during autumn environment often creates huge collection of waste each year. To maintain and manage a clean environment, it is necessary to collect the dried leaf waste regularly at rapid intervals. In order of claiming this, it is very essential to development a leaf falling simulator using objective detection principle that measures the size of the leaf, color and other relevant features for efficient feature selection, detection and collection of waste. In order to accommodate these three tasks, the study uses a machine learning detection that essentially identifies the leaves based on input datasets. The study considers various input features like size of the leaf, color, falling rate of a dried leaf and moment of inertia. These features are utilized for detecting the falling leaves and providing the input for clearing the leaf waste in that region. The experiments are conducted to test the real-time applicability of the model against various trees and in different regions.