In recent years, the estimation of tobacco field areas has become a critical component of precision tobacco cultivation. However, traditional satellite remote sensing methods face challenges such as high costs, low accuracy, and susceptibility to noise, making it difficult to meet the demand for high precision. Additionally, optical remote sensing methods perform poorly in regions with complex terrain. Therefore, Unmanned Aerial Vehicle multispectral remote sensing technology has emerged as a viable solution due to its high resolution and rich spectral information. This study employed a DJI Mavic 3M equipped with high-resolution RGB and multispectral cameras to collect tobacco field data covering five bands: RGB, RED, RED EDGE, NIR, and GREEN in Agang Town, Luoping County, Yunnan Province, China. To ensure the accuracy of the experiment, we used 337, 242, and 215 segmented tobacco field images for model training, targeting both RGB channels and seven-channel data. We developed a tobacco field semantic segmentation method based on PP-LiteSeg and deeply customized the model to adapt to the characteristics of multispectral images. The input layer’s channel number was adjusted to multiple channels to fully utilize the information from the multispectral images. The model structure included an encoder, decoder, and SPPM module, which used a multi-layer convolution structure to achieve feature extraction and segmentation of multispectral images. The results indicated that compared to traditional RGB images, multispectral images offered significant advantages in handling edges and complex terrain for semantic segmentation. Specifically, the predicted area using the seven-channel data was 11.43 m² larger than that obtained with RGB channels. Additionally, the seven-channel model achieved a prediction accuracy of 98.84%. This study provides an efficient and feasible solution for estimating tobacco field areas based on multispectral images, offering robust support for modern agricultural management.

The application of image recognition technology plays a vital role in agricultural disease identification. Existing approaches primarily rely on image classification, object detection, or semantic segmentation. However, a major challenge in current semantic segmentation methods lies in accurately identifying small target objects. In this study, common tobacco leaf diseases—such as frog-eye disease, climate spots, and wildfire disease—are characterized by small lesion areas, with an average target size of only 32 pixels. This poses significant challenges for existing techniques to achieve precise segmentation. To address this issue, we propose integrating two attention mechanisms, namely cross-feature map attention and dual-branch attention, which are incorporated into the semantic segmentation network to enhance performance on small lesion segmentation. Moreover, considering the lack of publicly available datasets for tobacco leaf disease segmentation, we constructed a training dataset via image splicing. Extensive experiments were conducted on baseline segmentation models, including UNet, DeepLab, and HRNet. Experimental results demonstrate that the proposed method improves the mean Intersection over Union (mIoU) by 4.75% on the constructed dataset, with only a 15.07% increase in computational cost. These results validate the effectiveness of our novel attention-based strategy in the specific context of tobacco leaf disease segmentation.

Accurate fine-scale forest mapping is fundamental for ecological monitoring and resource management. While deep learning semantic segmentation methods have advanced the interpretation of high-resolution UAV imagery, their generalization across diverse forest regions remains challenging due to high spatial heterogeneity. To address this, we propose two enhanced versions based on the PP-LiteSeg architecture for robust cross-regional forest segmentation. Version 01 (V01) integrates a multi-branch attention fusion module composed of parallel channel, spatial, and pixel attention branches. This design enables fine-grained feature enhancement and precise boundary delineation in structurally regular artificial forests, such as the Huayuan Forest Farm. As a result, V01 achieves a mIoU of 92.64% and an F1-score of 96.10%, representing an approximately 18 percentage-point mIoU improvement over PSPNet and DeepLabv3+. Building on this, Version 02 (V02) introduces a lightweight residual connection that directly shortcuts the fused features, thereby improving feature stability and robustness under complex textures and illumination, and demonstrates stronger performance in naturally heterogeneous forests (Longhai Township), attaining an mIoU of 91.87% and an F1-score of 95.77% (5.72 percentage-point mIoU gain over DeepLabv3+). We further conduct comprehensive comparisons against conventional CNN baselines as well as representative lightweight and transformer-based models (BiSeNetV2 and SegFormer-B0). In bidirectional cross-region transfer (train on one region and directly test on the other), V02 exhibits the most stable performance with minimal degradation, highlighting its robustness under domain shift. On a combined cross-regional dataset, V02 achieves a leading mIoU of 91.50%, outperforming U-Net, DeepLabv3+, and PSPNet. In summary, V01 excels in boundary delineation for regular plantation forests, whereas V02 shows more stable generalization across highly varied natural forest landscapes, providing practical solutions for region-adaptive UAV forest segmentation.