Bridging Classical and Modern Computer Vision: PerceptiveNet for Tree Crown Semantic Segmentation
This work addresses a domain-specific problem for forest management and environmental science, offering incremental advancements by integrating classical and modern computer vision techniques.
The paper tackled the challenge of accurate tree crown semantic segmentation in remote sensing data by proposing PerceptiveNet, a model that incorporates a Logarithmic Gabor-parameterised convolutional layer and a backbone with a wider receptive field, achieving significant performance improvements over state-of-the-art models on a tree crown dataset and generalizing across benchmark aerial scene datasets.
The accurate semantic segmentation of tree crowns within remotely sensed data is crucial for scientific endeavours such as forest management, biodiversity studies, and carbon sequestration quantification. However, precise segmentation remains challenging due to complexities in the forest canopy, including shadows, intricate backgrounds, scale variations, and subtle spectral differences among tree species. Compared to the traditional methods, Deep Learning models improve accuracy by extracting informative and discriminative features, but often fall short in capturing the aforementioned complexities. To address these challenges, we propose PerceptiveNet, a novel model incorporating a Logarithmic Gabor-parameterised convolutional layer with trainable filter parameters, alongside a backbone that extracts salient features while capturing extensive context and spatial information through a wider receptive field. We investigate the impact of Log-Gabor, Gabor, and standard convolutional layers on semantic segmentation performance through extensive experimentation. Additionally, we conduct an ablation study to assess the contributions of individual layers and their combinations to overall model performance, and we evaluate PerceptiveNet as a backbone within a novel hybrid CNN-Transformer model. Our results outperform state-of-the-art models, demonstrating significant performance improvements on a tree crown dataset while generalising across domains, including two benchmark aerial scene semantic segmentation datasets with varying complexities.