The western black-crested gibbon (Nomascus concolor) is a rare and endangered primate that inhabits southern China and northern Vietnam, and has become a key conservation target due to its distinctive call and highly endangered status, making its identification and monitoring particularly urgent. Identifying calls of the western black-crested gibbon using passive acoustic monitoring data is a crucial method for studying and analyzing these gibbons; however, traditional call recognition models often overlook the temporal information in audio features and fail to adapt to channel-feature weights. To address these issues, we propose an innovative deep learning model, VBSNet, designed to recognize and classify a variety of biological calls, including those of endangered western black-crested gibbons and certain bird species. The model incorporates the image feature extraction capability of the VGG16 convolutional network, the sequence modeling capability of bi-directional LSTM, and the feature selection capability of the SE attention module, realizing the multimodal fusion of image, sequence and attention information. In the constructed dataset, the VBSNet model achieved the best performance in the evaluation metrics of accuracy, precision, recall, and F1-score, realizing an accuracy of 98.35%, demonstrating high accuracy and generalization ability. This study provides an effective deep learning method in the field of automated bioacoustic monitoring, which is of great theoretical and practical significance for supporting wildlife conservation and maintaining biodiversity.

The estimation of Above-Ground Biomass (AGB) in Amorphophallus konjac (Konjac) is essential for field management and yield prediction. While previous research has demonstrated the efficacy of Unmanned Aerial Vehicle (UAV) RGB imagery in estimating AGB for monoculture crops, the applicability of these methods to AGB estimation in Konjac remains uncertain due to its distinct morphological traits and prevalent intercropping practices with maize. Additionally, the Vegetation Indices (VIs) and Texture Features (TFs) obtained from UAV-based RGB imagery exhibit significant redundancy, raising concerns about whether the selected optimal variables can maintain estimation accuracy. Therefore, this study assessed the effectiveness of Variable Selection Using Random Forests (VSURF) and Principal Component Analysis (PCA) in variable selection and compared the performance of Stepwise Multiple Linear Regression (SMLR) with four Machine Learning (ML) regression techniques: Random Forest Regression (RFR), Extreme Gradient Boosting Regression (XGBR), Partial Least Squares Regression (PLSR), and Support Vector Regression (SVR), as well as Deep Learning (DL), in estimating the AGB of Konjac based on the selected features. The results indicate that the integration (PCA_(PCA_VIs+PCA_TFs)) of PCA-based VIs and PCA-based TFs using PCA achieved the best prediction accuracy (R² = 0.96, RMSE = 0.08 t/hm², MAE = 0.06 t/hm²) with SVR. In contrast, the DL model derived from AlexNet, combined with RGB imagery, yielded moderate predictive accuracy (R² = 0.72, RMSE = 0.21 t/hm², MAE = 0.17 t/hm²) compared with the optimal ML model. Our findings suggest that ML regression techniques, combined with appropriate variable-selected approaches, outperformed DL techniques in estimating the AGB of Konjac. This study not only provides new insights into AGB estimation in Konjac but also offers valuable guidance for estimating AGB in other crops, thereby advancing the application of UAV technology in crop biomass estimation.

The endangered Gaoligong hoolock gibbon (Hoolock tianxing) faces significant threats from habitat degradation and loss, making accurate habitat assessment crucial for effective conservation. This study explored the effectiveness of high-resolution small unoccupied aerial system (sUAS) imagery for evaluating habitat quality, comparing its performance against Sentinel-2 satellite data. Focusing on the critically fragmented habitat of this primate in Yingjiang County, China, we aimed to (1) assess habitat quality at the patch level using a sUAS; (2) apply the InVEST Habitat Quality (IHQ) model; and (3) compare the effectiveness of sUAS and Sentinel-2 imagery, across different resolutions, for habitat quality evaluation. We utilized sUAS imagery (0.05 m resolution) obtained from a DJI Mavic 3 drone and Sentinel-2 data (10 m resolution) for a comparative analysis. The InVEST IHQ model was then used to analyze nine habitat patches, examining how data resolution impacts habitat quality assessments. Our results showed that habitat quality varied considerably across space, with lower quality observed near villages due to agricultural activity and infrastructure development. The sUAS imagery proved superior at capturing detailed landscape features and delineating small, fragmented patches compared to Sentinel-2. Furthermore, the sUAS achieved higher classification accuracy. Although both data sources indicated generally high habitat quality, Sentinel-2 tended to overestimate both habitat quality and degradation compared to the sUAS. High-resolution sUAS imagery therefore provides a clear advantage for detailed habitat quality assessment and targeted conservation planning, especially in fragmented landscapes. Integrating sUAS data with other remote sensing methods is essential to improve the protection of endangered primate habitats. This research emphasizes the value of sUAS for fine-scale habitat analysis, providing a strong scientific basis for developing targeted habitat restoration strategies and guiding conservation management.

As part of the ecosystem, the western black-crested gibbon (Nomascus concolor) is important for ecological sustainability. Calls are an important means of communication for gibbons, so accurately recognizing and categorizing gibbon calls is important for their population monitoring and conservation. Since a large amount of sound data will be generated in the process of acoustic monitoring, it will take a lot of time to recognize the gibbon calls manually, so this paper proposes a western black-crested gibbon call recognition network based on SA_DenseNet-LSTM-Attention. First, to address the lack of datasets, this paper explores 10 different data extension methods to process all the datasets, and then converts all the sound data into Mel spectrograms for model input. After the test, it is concluded that WaveGAN audio data augmentation method obtains the highest accuracy in improving the classification accuracy of all models in the paper. Then, the method of fusion of DenseNet-extracted features and LSTM-extracted temporal features using PCA principal component analysis is proposed to address the problem of the low accuracy of call recognition, and finally, the SA_DenseNet-LSTM-Attention western black-crested gibbon call recognition network proposed in this paper is used for recognition training. In order to verify the effectiveness of the feature fusion method proposed in this paper, we classified 13 different types of sounds and compared several different networks, and finally, the accuracy of the VGG16 model improved by 2.0%, the accuracy of the Xception model improved by 1.8%, the accuracy of the MobileNet model improved by 2.5%, and the accuracy of the DenseNet network model improved by 2.3%. Compared to other classical chirp recognition networks, our proposed network obtained the highest accuracy of 98.2%, and the convergence of our model is better than all the compared models. Our experiments have demonstrated that the deep learning-based call recognition method can provide better technical support for monitoring western black-crested gibbon populations.