Self-supervised Learning for Panoptic Segmentation of Multiple Fruit Flower Species

TL;DR

This paper introduces a self-supervised learning approach for panoptic segmentation of multiple fruit and flower species, improving accuracy without relying on manual labels or costly post-processing.

Contribution

It presents a novel self-supervised training strategy using pseudo-labels and data augmentation for multi-species flower segmentation in precision agriculture.

Findings

Outperforms state-of-the-art models on multi-species flower dataset

Eliminates need for computationally expensive post-processing

Provides a new baseline for flower detection applications

Abstract

Convolutional neural networks trained using manually generated labels are commonly used for semantic or instance segmentation. In precision agriculture, automated flower detection methods use supervised models and post-processing techniques that may not perform consistently as the appearance of the flowers and the data acquisition conditions vary. We propose a self-supervised learning strategy to enhance the sensitivity of segmentation models to different flower species using automatically generated pseudo-labels. We employ a data augmentation and refinement approach to improve the accuracy of the model predictions. The augmented semantic predictions are then converted to panoptic pseudo-labels to iteratively train a multi-task model. The self-supervised model predictions can be refined with existing post-processing approaches to further improve their accuracy. An evaluation on a multi-species fruit tree flower dataset demonstrates that our method outperforms state-of-the-art models without computationally expensive post-processing steps, providing a new baseline for flower detection applications.