Explore the vast range of titles available.

Crop diseases can significantly affect various aspects of crop cultivation, including crop yield, quality, production costs, and crop loss. The utilization of modern technologies such as image analysis via machine learning techniques enables early and precise detection of crop diseases, hence empowering farmers to effectively manage and avoid the occurrence of crop diseases. The proposed methodology involves the use of modified MobileNetV3Large model deployed on edge device for real-time monitoring of grape leaf disease while reducing computational memory demands and ensuring satisfactory classification performance. To enhance applicability of MobileNetV3Large, custom layers consisting of two dense layers were added, each followed by a dropout layer, helped mitigate overfitting and ensured that the model remains efficient. Comparisons among other models showed that the proposed model outperformed those with an average train and test accuracy of 99.66% and 99.42%, with a precision, recall, and F1 score of approximately 99.42%. The model was deployed on an edge device (Nvidia Jetson Nano) using a custom developed GUI app and predicted from both saved and real-time data with high confidence values. Grad-CAM visualization was used to identify and represent image areas that affect the convolutional neural network (CNN) classification decision-making process with high accuracy. This research contributes to the development of plant disease classification technologies for edge devices, which have the potential to enhance the ability of autonomous farming for farmers, agronomists, and researchers to monitor and mitigate plant diseases efficiently and effectively, with a positive impact on global food security.

Apple leaf diseases seriously damage the yield and quality of apples. Current apple leaf disease diagnosis methods primarily rely on human visual inspection, which often results in low efficiency and insufficient accuracy. Many computer vision algorithms have been proposed to diagnose apple leaf diseases, but most of them are designed to run on high-performance GPUs. This potentially limits their application in the field, in which mobile devices are expected to be used to perform computer vision-based disease diagnosis on the spot. In this paper, we propose a lightweight one-stage network, called the Mobile Ghost Attention YOLO network (MGA-YOLO), which enables real-time diagnosis of apple leaf diseases on mobile devices. We also built a dataset, called the Apple Leaf Disease Object Detection dataset (ALDOD), that contains 8,838 images of healthy and infected apple leaves with complex backgrounds, collected from existing public datasets. In our proposed model, we replaced the ordinary convolution with the Ghost module to significantly reduce the number of parameters and floating point operations (FLOPs) due to cheap operations of the Ghost module. We then constructed the Mobile Inverted Residual Bottleneck Convolution and integrated the Convolutional Block Attention Module (CBAM) into the YOLO network to improve its performance on feature extraction. Finally, an extra prediction head was added to detect extra large objects. We tested our method on the ALDOD testing set. Results showed that our method outperformed other state-of-the-art methods with the highest mAP of 89.3%, the smallest model size of only 10.34 MB and the highest frames per second (FPS) of 84.1 on the GPU server. The proposed model was also tested on a mobile phone, which achieved 12.5 FPS. In addition, by applying image augmentation techniques on the dataset, mAP of our method was further improved to 94.0%. These results suggest that our model can accurately and efficiently detect apple leaf diseases and can be used for real-time detection of apple leaf diseases on mobile devices.

Fixed-wing UAVs (FW-UAVs) are empowered to handle diverse civilian and military missions, but sensor failure scenarios are constantly rising. Rapid advancement in deep learning methods currently proposes state-of-the-art solutions for fault detection of UAVs. However, most recent deep learning-based detection models suffer from model size, high computational complexity, and high-power consumption, which are challenging for small-sized FW-UAVs with limited battery backup and computational power. Therefore, to overcome these problems, this article introduces a lightweight CNN model built on prior work combined with the LSTM-AM framework to obtain accurate fault detection of FW-UAVs with low power consumption and fast computations. First, lightweight CNN architecture aims to minimize computational complexity while maintaining high accuracy in fault detection. The LSTM model merged with Attention Mechanism (AM), allows the architecture to obtain temporal dependencies and concentrate on essential features for enhanced fault detection accuracy. The combined version of lightweight CNN, LSTM, and AM commits to more reliable and efficient fault detection in FW-UAV applications, improving UAV drones’ overall performance and safety.

The maturity of tobacco leaves plays a decisive role in tobacco production, affecting the quality of the leaves and production control. Traditional recognition of tobacco leaf maturity primarily relies on manual observation and judgment, which is not only inefficient but also susceptible to subjective interference. Particularly in complex field environments, there is limited research on in situ field maturity recognition of tobacco leaves, making maturity recognition a significant challenge. In response to this problem, this study proposed a MobileNetV1 model combined with a Feature Pyramid Network (FPN) and attention mechanism for in situ field maturity recognition of tobacco leaves. By introducing the FPN structure, the model fully exploits multi-scale features and, in combination with Spatial Attention and SE attention mechanisms, further enhances the expression ability of feature map channel features. The experimental results show that this model, with a size of 13.7 M and FPS of 128.12, performed outstandingly well on the task of field maturity recognition of tobacco leaves, achieving an accuracy of 96.3%, superior to classical models such as VGG16, VGG19, ResNet50, and EfficientNetB0, while maintaining excellent computational efficiency and small memory footprint. Experiments were conducted involving noise perturbations, changes in environmental brightness, and occlusions to validate the model’s robustness in dealing with the complex environments that may be encountered in actual applications. Finally, the Score-CAM algorithm was used for result visualization. Heatmaps showed that the vein and color variations of the leaves provide key feature information for maturity recognition. This indirectly validates the importance of leaf texture and color features in maturity recognition and, to some extent, enhances the credibility of the model. The model proposed in this study maintains high performance while having low storage requirements and computational complexity, making it significant for in situ field maturity recognition of tobacco leaves.

Falls are a major contributor to morbidity and mortality among older adults, and timely fall detection can help reduce the severity of fall-related outcomes. Wearable inertial measurement unit (IMU) sensors offer a promising solution for fall detection; however, many existing approaches rely on multiple sensing locations and computationally intensive models, which can limit their practicality for resource-constrained wearable devices. This study proposes a knowledge distillation framework for efficient wrist-based fall detection using the publicly available University of Málaga fall detection dataset (UMAFall), a benchmark dataset for human activity recognition and fall detection. Although UMAFall was not collected from older adults, it provides a useful public benchmark for evaluating IMU-based fall detection methods. Knowledge distillation was implemented using a teacher–student framework, in which a high-capacity teacher model trained with IMU data from four body locations (waist, wrist, ankle, and chest) provided soft targets for guiding a compact wrist-only CNN student model. In a held-out test evaluation using Subjects 2 and 5, the teacher model achieved 97.6% accuracy and an F1 score of 96.7%, with approximately 1.3 million trainable parameters. The independently trained wrist-based CNN achieved 90.2% accuracy and an F1 score of 87.1%. After applying knowledge distillation, the student model improved to 95.1% accuracy and an F1 score of 93.3% while maintaining the same lightweight architecture. A supplementary leave-one-subject-out analysis showed slightly higher and more stable AUC for KD-CNN than the independently trained CNN (0.96 ± 0.03 vs. 0.94 ± 0.07). These findings suggest that knowledge distillation can improve wrist-only fall detection in this feasibility evaluation, but further validation using older adults and real-world smartwatch data is needed.

White blood cells (WBC), which are human peripheral blood cells, are the most significant part of the immune system that defends the body against microorganisms. Modifications in the morphological structure and number of subtypes of WBC play an major role in the diagnosis of serious diseases such as anemia and leukemia. Therefore, accurate WBC classification is clinically quite significant in the diagnosis of the disease. In last years, deep learning, especially CNN, has been used frequently in the field of medicine because of its strong self-learning capabilities and it can extract deeper features in images with stronger semantic information. In this study, a new CNN-based method is proposed for WBC classification. The proposed method (PM) is a hybrid method consisting of Inception module, pyramid pooling module (PPM) and depthwise squeeze-and-excitation block (DSEB). Inception module increases classification accuracy of CNNs by performing multiple parallel convolutions at different scales. PPM captures multi-scale contextual information from the input image by pooling features at multiple different scales. DSEB offers a structure where the network can selectively learn about informative features and remove useless ones. For the analysis of the classification results of the PM, experiments were carried out on three different datasets consisting of four classes (BCCD dataset), five classes (Raabin WBC dataset) and eight classes. As a result of the experimental studies, classification accuracy was obtained 99.96% in the BCCD dataset containing 4 classes, 99.22% in the Raabin WBC dataset containing 5 classes and 99.72% in the PBC dataset containing 8 classes. Compared with the state-of-the-art studies in the literature, the PM achieved the best accuracy in three datasets.

Autonomous navigation in developing regions is challenged by heterogeneous traffic, dynamic occlusions, and weak road structure. Existing segmentation models, largely trained on structured Western datasets, struggle to generalize under these conditions. To address this gap, we propose IndiVNet, a semantic segmentation architecture tailored for unstructured Indian driving environments. IndiVNet introduces a progressive dilation encoder (6 16) that captures fine-grained details and broad contextual cues without inducing oversparsity. Evaluated on the India Driving Dataset (IDD), it achieves 69.98% mIoU, outperforming CNN and Transformer baselines, and reaches 73.2% mIoU on CAMVID, demonstrating strong cross-domain generalization. By combining contextual adaptability with real-time efficiency, IndiVNet offers a scalable, region-aware solution for robust autonomous navigation in complex environments.

Recent advancements in Convolutional Neural Networks (CNNs), combined with the growing adoption of farm-applicable Internet of Things (IoT) devices, have expanded the application of precision agriculture in mango orchards. The Smart Mango Orchard can play a crucial role in ensuring mango trees thrive and produce high-quality fruit. However, state-of-the-art (SOTA) CNNs are built on numerous layers and many parameters; therefore, they are challenging to deploy in IoT devices. However, the lightweight CNN is a possible solution. This study developed a lightweight CNN, mangoNet, to deploy in an innovative mango orchard environment. The mangoNet is expected to monitor the mango leaf images and report them to farmers via a mobile app using the IoT system. The study was conducted using the primary dataset collected from the mango gardens in Rajshahi, Bangladesh. The mangoNet benchmark was evaluated using six SOTA CNNs. The mangoNet, with only 3,987,400 Parameters, outperforms SOTA CNNs’ accuracy (99.61%). In addition, this study employed SHAP, LIME, and Grad-CAM visualizations to identify and depict the image regions that contribute to mangoNet’s decision-making process. The mangoNet is integrated into a Streamlit web application and an Android mobile app, as researchers suggest for the practical use of CNNs. The novelty of mangoNet lies in its balanced sequential architecture, with a careful selection of kernels and progressive filter expansion, enabling early layers to capture low-level features and deeper layers to extract high-level features. As a result, the proposed mangoNet achieved high accuracy while requiring fewer computational resources and reduced training time. In addition, the mangoNet-powered website and mobile application empower both farmers and farming stakeholders by making real-time disease detection. In the future, the prototype is expected to be commercialized as Bangladesh is the 8 Th mango-producing country in the world.

Loop Closure Detection (LCD) is an important technique to improve the accuracy of Simultaneous Localization and Mapping (SLAM). In this paper, we propose an LCD algorithm based on binary classification for feature matching between similar images with deep learning, which greatly improves the accuracy of LCD algorithm. Meanwhile, a novel lightweight convolutional neural network (CNN) is proposed and applied to the target detection task of key frames. On this basis, the key frames are binary classified according to their labels. Finally, similar frames are input into the improved lightweight feature matching network based on Transformer to judge whether the current position is loop closure. The experimental results show that, compared with the traditional method, LFM-LCD has higher accuracy and recall rate in the LCD task of indoor SLAM while ensuring the number of parameters and calculation amount. The research in this paper provides a new direction for LCD of robotic SLAM, which will be further improved with the development of deep learning.

Electroencephalography is a widely used non-invasive method for monitoring brain electrical activity, critical for diagnosing and managing neurological disorders such as epilepsy. While clinical standards use 21 electrodes to capture comprehensive neural signals, a personalized approach can enhance performance by selecting patient-specific channels, reducing noise and redundancy. This study introduces an innovative, lightweight deep learning system optimized for real-time seizure detection in personalized wearable devices. The system uses an efficient Convolutional Neural Network that processes data from just two channels. These channels are automatically selected using a data-driven mechanism that identifies the most informative scalp regions based on each patient’s unique seizure patterns. The proposed approach ensures high reliability, even with small datasets, and improves interpretability for clinicians by overcoming the limitations of more complex methods. The tailored channel selection boosts detection accuracy and ensures robust performance across different seizure types while reducing the computational burden typical of multi-electrode systems. Validation on the publicly available CHB-MIT dataset achieved an average balanced accuracy of 0.83 and a false-positive rate of approximately 0.1/h. The system’s performance matches, and in some cases outperforms, state-of-the-art systems that use four fixed channels in temporal regions, demonstrating the potential of two-channel wearable solutions, specifically with a non-negligible 30% reduction in the false-positive rate. This interpretable, patient-specific method enables the development of personalized, efficient, and compact wearable devices for reliable seizure detection in everyday life.