Explore the vast range of titles available.

Deep learning is widely used in modern orchard production for various inspection missions, which helps improve the efficiency of orchard operations. In the mission of visual detection during fruit picking, most current lightweight detection models are not yet effective enough to detect multi-type occlusion targets, severely affecting automated fruit-picking efficiency. This study addresses this problem by proposing the pioneering design of a multi-type occlusion apple dataset and an augmentation method of data balance. We divided apple occlusion into eight types and used the proposed method to balance the number of annotation boxes for multi-type occlusion apple targets. Finally, a validation experiment was carried out using five popular lightweight object detection models: yolox-s, yolov5-s, yolov4-s, yolov3-tiny, and efficidentdet-d0. The results show that, using the proposed augmentation method, the average detection precision of the five popular lightweight object detection models improved significantly. Specifically, the precision increased from 0.894 to 0.974, recall increased from 0.845 to 0.972, and mAP0.5 increased from 0.982 to 0.919 for yolox-s. This implies that the proposed augmentation method shows great potential for different fruit detection missions in future orchard applications.

Welding is the most expensive process in building ships and offshore plants. Therefore, the quantity of welding material should be calculated for the subsections (cells) of the blocks for efficient work planning, and welding paths must be generated for welding automation. Three-dimensional (3D) computer-aided design (CAD) models have been used for this work. However, relevant information regarding welding is often omitted, and a separate database and interface to this database must be developed. In this study, a method of lightweight model-based weld line generation is proposed, followed by the calculation of bead length for welding material quantity estimation and welding path generation. Experiments were performed on various test cases of curved parts and blocks. The proposed method accurately generated weld lines, calculated bead length, and generated welding paths in a short time of approximately 1 s.

Deep Neural Network (DNN) models for both resource-rich environments and resource-constrained devices have become abundant in recent years. As of now, the literature on different available options for the design, development, and deployment of DNN models to resource constrained-end devices is limited and demands extensive further study. This paper reviews vital research efforts for the design of DNN models while deploying them at the end devices such as smart cameras for real-time object detection tasks. The design ideas include the types of DNN models, hardware and software requirements for the development, resource constraints imposed by the computing devices, and the optimization techniques required for the efficient processing of DNN. The study also aims to conduct a systematic literature review on current trends in different real-time applications of DNN models and explores the following four dimensions: (1) DNN model perspective: to associate appropriate DNN models with the proper hardware to achieve optimal throughput. (2) Hardware perspective: to answer different available options in hardware platforms for achieving on-device intelligence. (3) Resources and optimization perspective: to analyze the type of resource limitations in hardware platforms and the use of optimization techniques to overcome the performance issues. (4) Application perspective: to understand the real-time uses of DNN models in different application domains. This work also explores different performance measures that need to be considered for on-device intelligence and provides possible future directions for the challenges reviewed.

With the development of deep learning technology, researchers are increasingly paying attention to how to efficiently salvage surface garbage. Since the 1980s, the development of plastic products and economic growth has led to the accumulation of a large amount of garbage in rivers. Due to the large amount of garbage and the high risk of surface operations, the efficiency of manual garbage retrieval will be greatly reduced. Among existing methods, using YOLO algorithm to detect target objects is the most popular. Compared to traditional detection algorithms, YOLO algorithm not only has higher accuracy, but also is more lightweight. This article presents a lightweight YOLOv5 water surface garbage detection algorithm suitable for deployment on unmanned ships. This article has been validated on the Orca dataset, experimental results showed that the detection speed of the improved YOLOv5 increased by 4.3%, mAP value reached 84.9%, precision reached 88.7%, the parameter quantity only accounts for 12% of the original data. Compared with the original algorithm, the improved algorithm not only has higher accuracy, but also can be applied to more hardware devices due to its lighter weight.

Currently, with the rapid development of deep learning, deep neural networks (DNNs) have been widely applied in various computer vision tasks. However, in the pursuit of performance, advanced DNN models have become more complex, which has led to a large memory footprint and high computation demands. As a result, the models are difficult to apply in real time. To address these issues, model compression has become a focus of research. Furthermore, model compression techniques play an important role in deploying models on edge devices. This study analyzed various model compression methods to assist researchers in reducing device storage space, speeding up model inference, reducing model complexity and training costs, and improving model deployment. Hence, this paper summarized the state-of-the-art techniques for model compression, including model pruning, parameter quantization, low-rank decomposition, knowledge distillation, and lightweight model design. In addition, this paper discusses research challenges and directions for future work.

Due to the constraints of agricultural computing resources and the diversity of plant diseases, it is challenging to achieve the desired accuracy rate while keeping the network lightweight. In this paper, we proposed a computationally efficient deep learning architecture based on the mobile vision transformer (MobileViT) for real-time detection of plant diseases, which we called plant-based MobileViT (PMVT). Our proposed model was designed to be highly accurate and low-cost, making it suitable for deployment on mobile devices with limited resources. Specifically, we replaced the convolution block in MobileViT with an inverted residual structure that employs a 7×7 convolution kernel to effectively model long-distance dependencies between different leaves in plant disease images. Furthermore, inspired by the concept of multi-level attention in computer vision tasks, we integrated a convolutional block attention module (CBAM) into the standard ViT encoder. This integration allows the network to effectively avoid irrelevant information and focus on essential features. The PMVT network achieves reduced parameter counts compared to alternative networks on various mobile devices while maintaining high accuracy across different vision tasks. Extensive experiments on multiple agricultural datasets, including wheat, coffee, and rice, demonstrate that the proposed method outperforms the current best lightweight and heavyweight models. On the wheat dataset, PMVT achieves the highest accuracy of 93.6% using approximately 0.98 million (M) parameters. This accuracy is 1.6% higher than that of MobileNetV3. Under the same parameters, PMVT achieved an accuracy of 85.4% on the coffee dataset, surpassing SqueezeNet by 2.3%. Furthermore, out method achieved an accuracy of 93.1% on the rice dataset, surpassing MobileNetV3 by 3.4%. Additionally, we developed a plant disease diagnosis app and successfully used the trained PMVT model to identify plant disease in different scenarios.

The complex backgrounds of crop disease images and the small contrast between the disease area and the background can easily cause confusion, which seriously affects the robustness and accuracy of apple disease- identification models. To solve the above problems, this paper proposes a Vision Transformer-based lightweight apple leaf disease- identification model, ConvViT, to extract effective features of crop disease spots to identify crop diseases. Our ConvViT includes convolutional structures and Transformer structures; the convolutional structure is used to extract the global features of the image, and the Transformer structure is used to obtain the local features of the disease region to help the CNN see better. The patch embedding method is improved to retain more edge information of the image and promote the information exchange between patches in the Transformer. The parameters and FLOPs (Floating Point Operations) of the model are significantly reduced by using depthwise separable convolution and linear-complexity multi-head attention operations. Experimental results on a complex background of a self-built apple leaf disease dataset show that ConvViT achieves comparable identification results (96.85%) with the current performance of the state-of-the-art Swin-Tiny. The parameters and FLOPs are only 32.7% and 21.7% of Swin-Tiny, and significantly ahead of MobilenetV3, Efficientnet-b0, and other models, which indicates that the proposed model is indeed an effective disease-identification model with practical application value.

This paper addresses the industrial demand for precision and efficiency in metal surface defect detection by proposing SLF-YOLO, a lightweight object detection model designed for resource-constrained environments. The key innovations of SLF-YOLO include a novel SC_C2f module with a channel gating mechanism to enhance feature representation and regulate information flow, and a newly designed Light-SSF_Neck structure to improve multi-scale feature fusion and morphological feature extraction. Additionally, an improved FIMetal-IoU loss function is introduced to boost generalization performance, particularly for fine-grained and small-target defects. Experimental results demonstrate that SLF-YOLO achieves a mean Average Precision (mAP) of 80.0% on the NEU-DET dataset, outperforming YOLOv8’s 75.9%. On the AL10-DET dataset, SLF-YOLO achieves a mAP of 86.8%, striking an effective balance between detection accuracy and computational efficiency without increasing model complexity. Compared to other mainstream models, SLF-YOLO demonstrates strong detection accuracy while maintaining a lightweight architecture, making it highly suitable for industrial applications in metal surface defect detection. The source code is available at https://github.com/zacianfans/SLF-YOLO .

Aerial image target detection is essential for urban planning, traffic monitoring, and disaster assessment. However, existing detection algorithms struggle with small target recognition and accuracy in complex environments. To address this issue, this paper proposes an improved model based on YOLOv8, named MPE-YOLO. Initially, a multilevel feature integrator (MFI) module is employed to enhance the representation of small target features, which meticulously moderates information loss during the feature fusion process. For the backbone network of the model, a perception enhancement convolution (PEC) module is introduced to replace traditional convolutional layers, thereby expanding the network’s fine-grained feature processing capability. Furthermore, an enhanced scope-C2f (ES-C2f) module is designed, utilizing channel expansion and stacking of multiscale convolutional kernels to enhance the network’s ability to capture small target details. After a series of experiments on the VisDrone, RSOD, and AI-TOD datasets, the model has not only demonstrated superior performance in aerial image detection tasks compared to existing advanced algorithms but also achieved a lightweight model structure. The experimental results demonstrate the potential of MPE-YOLO in enhancing the accuracy and operational efficiency of aerial target detection. Code will be available online (https://github.com/zhanderen/MPE-YOLO).

Accurate detection of transportation objects is pivotal for enhancing driving safety and operational efficiency. In the rapidly evolving domain of transportation systems, the utilization of unmanned aerial vehicles (UAVs) for low-altitude detection, leveraging remotely-sensed images and videos, has become increasingly vital. Addressing the growing demands for robust, real-time object-detection capabilities, this study introduces a lightweight, memory-efficient model specifically engineered for the constrained computational and power resources of UAV-embedded platforms. Incorporating the FasterNet-16 backbone, the model significantly enhances feature-processing efficiency, which is essential for real-time applications across diverse UAV operations. A novel multi-scale feature-fusion technique is employed to improve feature utilization while maintaining a compact architecture through passive integration methods. Extensive performance evaluations across various embedded platforms have demonstrated the model’s superior capabilities and robustness in real-time operations, thereby markedly advancing UAV deployment in crucial remote-sensing tasks and improving productivity and safety across multiple domains.