Plantation Monitoring Using Drone Images: A Dataset and Performance Review

TL;DR

This paper introduces a drone image dataset for plantation health monitoring, evaluates CNN models on it, and explores depth-wise convolution and object detection to improve accuracy in assessing tree health.

Contribution

It presents a new drone image dataset with annotations for tree health categories and analyzes CNN model performance, highlighting the benefits of depth-wise convolution and object detection.

Findings

Depth-wise convolution improves model accuracy.

Object detection helps identify individual trees.

Initial models show low accuracy, indicating dataset complexity.

Abstract

Automatic monitoring of tree plantations plays a crucial role in agriculture. Flawless monitoring of tree health helps farmers make informed decisions regarding their management by taking appropriate action. Use of drone images for automatic plantation monitoring can enhance the accuracy of the monitoring process, while still being affordable to small farmers in developing countries such as India. Small, low cost drones equipped with an RGB camera can capture high-resolution images of agricultural fields, allowing for detailed analysis of the well-being of the plantations. Existing methods of automated plantation monitoring are mostly based on satellite images, which are difficult to get for the farmers. We propose an automated system for plantation health monitoring using drone images, which are becoming easier to get for the farmers. We propose a dataset of images of trees with three categories: ``Good health", ``Stunted", and ``Dead". We annotate the dataset using CVAT annotation tool, for use in research purposes. We experiment with different well-known CNN models to observe their performance on the proposed dataset. The initial low accuracy levels show the complexity of the proposed dataset. Further, our study revealed that, depth-wise convolution operation embedded in a deep CNN model, can enhance the performance of the model on drone dataset. Further, we apply state-of-the-art object detection models to identify individual trees to better monitor them automatically.