Dual Atrous Separable Convolution for Improving Agricultural Semantic Segmentation

TL;DR

This paper introduces a novel Dual Atrous Separable Convolution module integrated into a DeepLabV3 framework, significantly improving agricultural image segmentation efficiency and accuracy while maintaining low computational complexity.

Contribution

The study presents a new DAS Conv module and a strategic skip connection, enhancing segmentation performance and efficiency in agricultural imagery compared to existing models.

Findings

Outperforms baseline models in agricultural segmentation tasks.

Achieves over 66% efficiency improvement over state-of-the-art models.

Maintains high segmentation quality with lower computational cost.

Abstract

Agricultural image semantic segmentation is a pivotal component of modern agriculture, facilitating accurate visual data analysis to improve crop management, optimize resource utilization, and boost overall productivity. This study proposes an efficient image segmentation method for precision agriculture, focusing on accurately delineating farmland anomalies to support informed decision-making and proactive interventions. A novel Dual Atrous Separable Convolution (DAS Conv) module is integrated within the DeepLabV3-based segmentation framework. The DAS Conv module is meticulously designed to achieve an optimal balance between dilation rates and padding size, thereby enhancing model performance without compromising efficiency. The study also incorporates a strategic skip connection from an optimal stage in the encoder to the decoder to bolster the model's capacity to capture fine-grained spatial features. Despite its lower computational complexity, the proposed model outperforms its baseline and achieves performance comparable to highly complex transformer-based state-of-the-art (SOTA) models on the Agriculture Vision benchmark dataset. It achieves more than 66% improvement in efficiency when considering the trade-off between model complexity and performance, compared to the SOTA model. This study highlights an efficient and effective solution for improving semantic segmentation in remote sensing applications, offering a computationally lightweight model capable of high-quality performance in agricultural imagery.