Explore the vast range of titles available.

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that severely affects memory, behavior, and cognitive function. Early and accurate diagnosis is crucial for effective intervention, yet detecting subtle changes in the early stages remains a challenge. In this study, we propose a hybrid deep learning-based multi-class classification system for AD using magnetic resonance imaging (MRI). The proposed approach integrates an improved DeepLabV3+ (IDeepLabV3+) model for lesion segmentation, followed by feature extraction using the LeNet-5 model. A novel feature selection method based on average correlation and error probability is employed to enhance classification efficiency. Finally, an Enhanced ResNext (EResNext) model is used to classify AD into four stages: non-dementia (ND), very mild dementia (VMD), mild dementia (MD), and moderate dementia (MOD). The proposed model achieves an accuracy of 98.12%, demonstrating its superior performance over existing methods. The area under the ROC curve (AUC) further validates its effectiveness, with the highest score of 0.97 for moderate dementia. This study highlights the potential of hybrid deep learning models in improving early AD detection and staging, contributing to more accurate clinical diagnosis and better patient care.

Accurate and efficient detection of landslide hazards is recognized as a critical requirement for both disaster emergency response and long-term land-use planning. However, conventional semantic segmentation models still present notable limitations when processing high-resolution remote sensing imagery, such as imprecise delineation of landslide boundaries and low performance in detecting small-scale landslides, often resulting in missed or false detections. To address these challenges, this study proposes FCA-DeepLab, a novel landslide detection model based on multimodal data fusion and an improved DeepLabv3 + architecture. The model integrates a multimodal fusion mechanism to achieve deep coupling of optical imagery and topographic features, thereby fully exploiting both visual and geomorphological contextual information related to landslides. Moreover, the conventional ResNet backbone is replaced with a ConvNeXt network employing 7 × 7 convolutional kernels, which substantially enlarges the receptive field and improves the ability to capture fine-grained features. A small‑object attention mechanism specifically designed for small targets is introduced to enhance sensitivity to subtle landslide characteristics and markedly reduce the missed detection rate. Comparative experiments on several public datasets demonstrate that FCA-DeepLab surpasses established semantic segmentation models such as UNet, Swin Transformer, SegFormer, and the original DeepLabv3 + in terms of overall accuracy, recall, and qualitative segmentation performance. Furthermore, additional evaluation on the Bijie landslide dataset confirms the model’s strong generalization capability, showing adaptability to diverse regions, complex terrains, and varied scenarios. These findings substantiate the proposed method’s significant advantages in improving detection accuracy, reducing false positives, and strengthening the identification of small-scale landslides, thereby providing a reliable technical reference for deep learning-based intelligent landslide detection.

Accurately identifying weeds in crop fields is key to achieving selective herbicide spraying. Weed identification is made difficult by the dense distribution of weeds and crops, which makes boundary segmentation at the overlap inaccurate, and thus pixels cannot be correctly classified. To solve this problem, this study proposes a soybean field weed recognition model based on an improved DeepLabv3+ model, which uses a Swin transformer as the feature extraction backbone to enhance the model’s utilization of global information relationships, fuses feature maps of different sizes in the decoding section to enhance the utilization of features of different dimensions, and adds a convolution block attention module (CBAM) after each feature fusion to enhance the model’s utilization of focused information in the feature maps, resulting in a new weed recognition model, Swin-DeepLab. Using this model to identify a dataset containing a large number of densely distributed weedy soybean seedlings, the average intersection ratio reached 91.53%, the accuracy improved by 2.94% compared with that before the improvement with only a 48 ms increase in recognition time, and the accuracy was superior to those of other classical semantic segmentation models. The results showed that the Swin-DeepLab network proposed in this paper can successfully solve the problems of incorrect boundary contour recognition when weeds are densely distributed with crops and incorrect classification when recognition targets overlap, providing a direction for the further application of transformers in weed recognition.

This paper proposes an improved Deeplabv3+ model for semantic segmentation of urban scenes targeting autonomous driving applications. A high-quality semantic segmentation dataset is constructed from 2,967 manually labeled aerial images captured at 200m height with a 5-eye camera. The images contain 5 classes - buildings, vegetation, ground, lake and playgrounds. The improved Deeplabv3+ network enriches high-level semantics by replacing max pooling with depthwise separable convolutions. Dilated convolutions extract multi-scale features to avoid overfitting. Experiments demonstrate that the model achieves an overall mean IoU of 0.87 on the test set, with IoU scores of 0.90, 0.92 and 0.94 on buildings, vegetation and water respectively. The model shows promising results for extracting semantic information from complex urban environments to support navigation for autonomous vehicles.

Aiming to determine the inaccurate image segmentation of strawberries with varying maturity levels due to fruit adhesion and stacking, this study proposed a strawberry image segmentation method based on the improved DeepLabV3+ model. The technique introduced the attention mechanism into the backbone network and the atrous spatial pyramid pooling module of the DeepLabV3+ network, adjusted the weights of feature channels in the neural network propagation process through the attention mechanism to enhance the feature information of strawberry images, reduced the interference of environmental factors, and improved the accuracy of strawberry image segmentation. The experimental results showed that the proposed method can accurately segment images of strawberries with different maturities; the mean pixel accuracy and mean intersection over union of the model were 90.9% and 83.05%, respectively, and the frames per second (FPS) was 7.67. The method can effectively reduce the influence of environmental factors on strawberry image segmentation and provide an effective approach for accurate operation of strawberry picking robots.

The accurate and rapid acquisition of fruit tree canopy parameters is fundamental for achieving precision operations in orchard robotics, including accurate spraying and precise fertilization. In response to the issue of inaccurate citrus tree canopy segmentation in complex orchard backgrounds, this paper proposes an improved DeepLabv3+ model for fruit tree canopy segmentation, facilitating canopy parameter calculation. The model takes the RGB-D (Red, Green, Blue, Depth) image segmented canopy foreground as input, introducing Dilated Spatial Convolution in Atrous Spatial Pyramid Pooling to reduce computational load and integrating Convolutional Block Attention Module and Coordinate Attention for enhanced edge feature extraction. MobileNetV3-Small is utilized as the backbone network, making the model suitable for embedded platforms. A citrus tree canopy image dataset was collected from two orchards in distinct regions. Data from Orchard A was divided into training, validation, and test set A, while data from Orchard B was designated as test set B, collectively employed for model training and testing. The model achieves a detection speed of 32.69 FPS on Jetson Xavier NX, which is six times faster than the traditional DeepLabv3+. On test set A, the mIoU is 95.62%, and on test set B, the mIoU is 92.29%, showing a 1.12% improvement over the traditional DeepLabv3+. These results demonstrate the outstanding performance of the improved DeepLabv3+ model in segmenting fruit tree canopies under different conditions, thus enabling precise spraying by orchard spraying robots.

This paper proposes an improved DeepLabV3+ lightweight algorithm for the identification of installation gaps in spacer replacement robots. By using lightweight MobileNetV3 to extract semantic features of spacer installation gaps, parameters and computational complexity are reduced; Perform dimensionality reduction and dimensionality increase operations on the ASPP module to reduce the number of model parameters; Introduce ECA module to restore the clarity of target boundaries; Use a loss function combining Focal Loss and Dice Loss to enhance segmentation performance. The experimental results show that the improved DeepLabV3+ algorithm improves MIoU, MPA, and prediction speed, while balancing segmentation accuracy and speed, and can effectively segment the installation gap of the spacer.

As the banana industry develops, the demand for intelligent banana crown cutting is increasing. To achieve efficient crown cutting of bananas, accurate segmentation of the banana crown is crucial for the operation of a banana crown cutting device. In order to address the existing challenges, this paper proposed a method for segmentation of banana crown based on improved DeepLabv3+. This method replaces the backbone network of the classical DeepLabv3+ model with MobilenetV2, reducing the number of parameters and training time, thereby achieving model lightweightness and enhancing model speed. Additionally, the Atrous Spatial Pyramid Pooling (ASPP) module is enhanced by incorporating the Shuffle Attention Mechanism and replacing the activation function with Meta-ACONC. This enhancement results in the creation of a new feature extraction module, called Banana-ASPP, which effectively handles high-level features. Furthermore, Multi-scale Channel Attention Module (MS-CAM) is introduced to the Decoder to improve the integration of features from multiple semantics and scales. According to experimental data, the proposed method has a Mean Intersection over Union (MIoU) of 85.75%, a Mean Pixel Accuracy (MPA) of 91.41%, parameters of 5.881 M and model speed of 61.05 f/s. Compared to the classical DeepLabv3+ network, the proposed model exhibits an improvement of 1.94% in MIoU and 1.21% in MPA, while reducing the number of parameters by 89.25% and increasing the model speed by 47.07 f/s. The proposed method enhanced banana crown segmentation accuracy while maintaining model lightweightness and speed. It also provided robust technical support for relevant parameters calculation of banana crown and control of banana crown cutting equipment.

It is necessary to develop automatic picking technology to improve the efficiency of litchi picking, and the accurate segmentation of litchi branches is the key that allows robots to complete the picking task. To solve the problem of inaccurate segmentation of litchi branches under natural conditions, this paper proposes a segmentation method for litchi branches based on the improved DeepLabv3+, which replaced the backbone network of DeepLabv3+ and used the Dilated Residual Networks as the backbone network to enhance the model’s feature extraction capability. During the training process, a combination of Cross-Entropy loss and the dice coefficient loss was used as the loss function to cause the model to pay more attention to the litchi branch area, which could alleviate the negative impact of the imbalance between the litchi branches and the background. In addition, the Coordinate Attention module is added to the atrous spatial pyramid pooling, and the channel and location information of the multi-scale semantic features acquired by the network are simultaneously considered. The experimental results show that the model’s mean intersection over union and mean pixel accuracy are 90.28% and 94.95%, respectively, and the frames per second (FPS) is 19.83. Compared with the classical DeepLabv3+ network, the model’s mean intersection over union and mean pixel accuracy are improved by 13.57% and 15.78%, respectively. This method can accurately segment litchi branches, which provides powerful technical support to help litchi-picking robots find branches.

The traditional carrying capacity assessment method fails to effectively quantify the difference in spatial distribution of rooftop photovoltaic (PV) resources and ignores the temporal fluctuation of PV output and load demand, as well as the temporal and spatial matching characteristics of sources and loads. This leads to problems such as a disconnect between the assessment and the actual grid acceptance capacity and insufficient dynamic adaptability. In response to the above issues, this paper proposes a dynamic assessment model for distributed photovoltaic carrying capacity based on the combination of improved DeepLabv3+ and game-theoretic weighted assignment. First, the DeepLabv3+ model was improved by integrating the Efficient Channel Attention (ECA) mechanism and the strip pooling (SP) module to enhance roof recognition accuracy. Ablation experiments showed that the mIoU increased to 77.53%, 6.29% higher than the original model. The simulation results in the summer scenario demonstrated that, with the optimal coordination of STMF and scene scoring, the comprehensive carrying coefficient reached 0.73. Next, a photovoltaic carrying capacity evaluation system was established, considering the source, grid, and load perspectives, with dynamic evaluation using a game-theory-based weighting method. Finally, a comprehensive carrying coefficient was introduced, accounting for the spatiotemporal match between photovoltaic output and load, leading to the development of a distributed photovoltaic carrying capacity model. The case study results show that, in summer, due to the optimal coordination of STMF and scene scoring, the comprehensive carrying coefficient reaches 0.73. With a total PV access capacity of 6.48 MW, all node voltages remain within limits, verifying the model’s effectiveness in grid adaptability.