Abstract:The study employs the mold hot pressing encapsulation method to embed fiber Bragg grating (FBG) sensors into a miniature glass fiber-reinforced polymer (GFRP) composite. Experimental results confirmed the feasibility of using miniature GFRP-FBG sensors to monitor the epoxy resin curing process. Additionally, the sensors exhibited excellent linearity and repeatability in load monitoring. These findings offer important insights for optimizing the encapsulation process and further advancing the use of GFRP-FBG sensors in material monitoring applications.

Abstract:In this paper, a novel convolutional neural network (CNN) assisted decoding method is proposed to recover information directly for underwater orbital angular momentum (OAM) multiplexing optical communication. The effects of various attenuations and ocean water types, such as absorption, scattering, turbulence fading, noise and diffraction, are considered comprehensively in our analysis. A regularly spaced continuous phase screen is used to represent ocean turbulence. And the angular diffraction function is exploited for simulating the propagation of the OAM beams. In order to minimize the bit error rate (BER) and simplify the receiver design, a CNN assisted decoding method is used to compensate the distorted OAM light and decode the transmission data directly without channel estimation and equalization. The CNN is trained to learn the multiplexed OAM light intensity map generated under various water environments. The bit error performance of CNN OAM system is also compared with that of traditional Gerchberg-Saxton (GS) algorithm. Our numerical simulation results indicate that the CNN assisted method combats the impairing effects of fading and improves the underwater OAM system performance obviously. Furthermore, it outperforms GS algorithm in almost all the turbulence environments at the same water environment. And the BER of the CNN assisted system still decreases effectively by increasing signal-to-noise ratio (SNR) even in moderate and strong turbulence situations while at the same time requiring less computation complexity.

Abstract:We propose a multi-line laser simulation system utilizing computer graphics and physical simulation to generate virtual multi-line laser datasets. Our framework provides key physical properties of the scene, including camera parameters, depth values, surface normals, and the actual two-dimensional (2D) and three-dimensional (3D) coordinates of the laser stripe centers for each rendered image. Beyond, we construct a virtual line laser scanning image dataset with a complex background by simulating interactions between lasers and object surfaces with the Monte Carlo method. With the proposed framework and dataset, a multi-line laser extraction algorithm based on a robust sorting algorithm is proposed and tested, which utilizes distance-based error analysis, connected component labeling, and iterative optimization refinement techniques. Both simulation and actual experiments show that our method outperforms the other state-of-the-art multi-line laser stripe center extraction methods. The proposed framework can be applied to different types of laser scanning systems in the future.

Abstract:Federated learning (FL), as a distributed learning paradigm, allows multiple medical institutions to collaborate on learning without the need to centralize all client data. However, existing methods pay little attention to more challenging medical image semantic segmentation tasks, especially in the scenario of the imbalanced dataset in federated few-shot learning (FSL). In this paper, we propose a subnetwork-based federated few-shot organ image segmentation method. Firstly, individual clients train using local training samples and then upload local model gradients to the server. The server utilizes their respective local model gradients to update the subnetwork maintained on the server and generate aggregation weights for forming personalized model parameters. Through this method, we can learn the similarities between different clients to address data heterogeneity issues. In addition, to enhance the communication efficiency between clients and the server, we have also designed a personalized layer aggregation strategy, which only transmits partial layer model parameters during the communication process to improve communication efficiency. Finally, we conducted experiments on abdomen magnetic resonance imaging (ABD-MRI) and abdomen computed tomography (ABD-CT) datasets to demonstrate the effectiveness of our method.

Abstract:With the increasing demand for traffic sign detection, the challenge of small target detection has become particularly prominent. The present study proposes an innovative approach by integrating knowledge distillation, L2 loss function, and convolutional block attention module (CBAM) mechanism to effectively tackle this issue. This series of improvements not only provide a new idea for small target detection, but also bring significant performance improvement in actual traffic scenes. Then, the integration method of the bidirectional feature pyramid network (BiFPN) is used to enhance the flexibility of the neural network to deal with input of different scales, while speeding up and improving the process of feature fusion. The experimental results demonstrate that when processing the Chinese city traffic sign detection benchmark (CCTSDB) dataset and executing the FLOW-IMG small target detection task, the optimized algorithm shows obvious performance improvement, and its accurate recognition rate jumps to 97% and 84.9%, respectively. For the basic algorithm, two datasets achieved improved accuracy by an innovative approach, improving accuracy by 5.8% and 1.3%, respectively. In terms of resource efficiency, compared to the original teacher model, the newly constructed model reduced the number of computing participants by approximately 15% during execution, while successfully reducing the overall computing task load by 14%.

Abstract:With the rapid development and wide application of internet of things (IoT) technology, optical equipment is being promoted to collect and store multi-scale remote sensing images, and to apply them in various fields such as industry, agriculture, and ecological and environmental protection. However, the resulting security risks have also caused widespread concern. This paper designs a remote sensing image synchronization encryption scheme based on 3D cross-coupled chaotic map and block-based cubes for multi-scale remote sensing images. First, a new 3D chaotic map is constructed by coupling the traditional single-node sinusoidal map, which provides support for building complex chaotic mappings for devices with limited resources. Second, a fractal square based on the Hilbert curve combined with chaos achieves simultaneous pixel scrambling and diffusion operations, improving multi-scale graphics encryption. Finally, a multi-type remote sensing graphics dataset is used for testing and a series of analysis experiments are performed to prove the feasibility of the algorithm.

Abstract:In the field of object detection, it is challenging to achieve a balance between neck complexity and accuracy. To address this issue, we propose an efficient and decoupled neck module called shared feature pyramid network (Shared-FPN). Not only does Shared-FPN not increase the number of model parameters and floating point operations per second (FLOPs), but it also can be easily ported to any of the detection models. It improves on path aggregation feature pyramid network (PAFPN) by using transposed convolution with a large convolution kernel as the upsampling module, designing spatial pyramidal pooling-fast downsampling (SPPFD) based on shared pooling, and designing shared convolution as the right part module. To evaluate the performance of Shared-FPN in object detection tasks, we conducted experiments on object detection datasets. The results show that Shared-FPN achieved excellent performance across all sizes. In particular, on the VOC 2012 dataset, the Shared-FPN’s mean average precision (mAP) was improved by 11.2% compared to FPN with you only look once extended-s (YOLOX-s) as the detector. On the COCO dataset, the Shared-FPN’s mAP was improved by 7.8% compared to FPN and 7.1% compared to PAFPN with faster region-based convolutional neural network (Faster RCNN) as the detector. The Shared-FPN can be easily inserted into any of the detectors for better performance in various scenarios, such as small, medium or large objects.

Abstract:Many studies based on convolutional neural networks (CNNs) for breast cancer axillary lymph node (ALN) images have focused on large sample analysis and clinical parameter integration, while limited attention has been paid to lightweight models for small ALN datasets. In this paper, we have selected a small number of ALN ultrasound image datasets as the research subject and designed a TX-GGCA model, consisting of the Tiny-Xception model and the global grouping coordinate attention (GGCA). The TX-GGCA demonstrated an accuracy of 99.14% and an area under curve (AUC) of 0.999 7 in classifying normal and abnormal ALN images, outperforming the best traditional model (accuracy:95.69%, AUC:0.993 2). It showed the potential value of this model for clinical diagnosis in primary hospitals with limited sample sizes.

Abstract:In this work, a material recognition technology based on the backscattering field of a target with vortex beam illumination is proposed to meet the application requirements of target material recognition and classification in laser detection. Firstly, the characteristics of the backscattering light field of the target with vortex beam illumination are analyzed, and it is proved that the spatial frequency bandwidth of the specklegram increases with the increase of the topological charge of the vortex beam, so that the features of the specklegram will become more abundant. Subsequently, an experimental setup was built to record the backscattering specklegram and establish a dataset for validation. Six typical artificial neural networks (ANNs) were used to achieve the task of target material recognition. With a dataset of 1 000 samples for each of three categories, the recognition accuracy can be up to 96.89%. Finally, a comprehensive evaluation model is established when we consider the factors, including recognition accuracy, model complexity, and training time, and the performances of these ANNs are compared. Among these ANNs, ResNet-18 exhibits superior overall performance. The proposed target material recognition technique paves a new way to multi-dimensional laser detection technology.

Abstract:This paper suggests an improved you only look once version 8n (YOLOv8n) algorithm for apple leaf disease detection, abbreviated as ALWB-YOLOv8n. The model is comprised of four essential components. Initially, arbitrary kernel convolution (AKConv) replaces the convolution module, which significantly decreases both the model’s parameter count and its overall size. Secondly, the large selective kernel network (LSKNet) attention mechanism is added in the Backbone, which can dynamically adjust the spatial sensory domain, and experiments have proved that this method is extremely advantageous for small target detection. Third, a weighted bi-directional feature pyramid network is introduced, which enables the model to achieve multi-scale feature fusion and is more concise and faster. Finally, wise intersection over union (WIoU) is used to replace complete intersection over union (CIoU) in YOLOv8, and the idea of focal loss is introduced, which effectively solves the detection problems in cases such as apple leaves occluding each other and blurred boundaries of diseased leaves. The improved algorithm exhibits superior performance compared to other common object detection algorithms. Compared with YOLOv8n, the improved algorithm achieves 2.3% improvement in precision, 3.8% improvement in recall, and 2.5% and 2.7% improvement in mAP0.5 and mAP0.5:0.95, respectively. Compared with YOLOv8n, the improved model reduces the number of parameters and size of the model and realizes real-time monitoring with a frames per second (FPS) of 50.5.