Image Quality Predictor with Highly Efficient Fully Convolutional Neural Network by Cao Yu-Dong, Liao Xin-Lin, Liu Hai-Yan

“…The parameters of the shared layers and the quality module are optimized firstly, and then, the parameters of the weight module are optimized with the designed loss function. …”
Get full text

Adaptive building engineering component extraction model based on DSOD by Na Lv, Xuan Yang

“…With the purpose to bring up the extraction efficiency and accuracy of building construction image component information, the dense block structure and loss function were proposed to optimize the deep supervised object detection algorithm, and an adaptive building construction component extraction model based on this algorithm was constructed. …”
Get full text

Real-Time Recognition and Localization of Kiwifruit Based on Improved YOLOv5s Algorithm by Jin-Sui Dai, Zhi-Qin He

“…First, data enhancement is implemented in the training strategy to ensure the model’s efficient learning and generalization ability in complex environments; then the Coordinate Attention mechanism is introduced to highlight the key features of the image and improve detail detection; Secondly, the Bidirectional Feature Pyramid Network structure is used to optimize the feature fusion and strengthen the information exchange between different layers through bidirectional connectivity; then, the loss function is optimized to improve the bounding box localization accuracy; and finally, the combined with the binocular vision stereo matching algorithm Semi-Global Block Matching to obtain the spatial location information of Kiwifruit. …”
Get full text

Label-Guided Data Augmentation for Chinese Named Entity Recognition by Miao Jiang, Honghui Chen

“…By employing twin encoders and a cross-attention mechanism, the model fuses sample and label semantics, while the optimized loss function adapts to class imbalances. …”
Get full text

Audio recognition method of belt conveyor roller fault based on convolutional neural network and linear regression by Xiangyuan CHEN, Wei QIN, Yanchi LIU, Minghua LUO

“…The sample weight of the model training is optimized by the Focal Loss loss function to improve the accuracy of the model for roller fault recognition.The method in this paper is used to analyze and verify the audio information of the fault roller of the belt conveyor actually collected in Guojiawan Coal Mine of Yulin. …”
Get full text

YOLOv8n-SMMP: A Lightweight YOLO Forest Fire Detection Model by Nianzu Zhou, Demin Gao, Zhengli Zhu

“…Key innovations include the following: introducing the SlimNeck solution to streamline the neck network by replacing conventional convolutions with Group Shuffling Convolution (GSConv) and substituting the Cross-convolution with 2 filters (C2f) module with the lightweight VoV-based Group Shuffling Cross-Stage Partial Network (VoV-GSCSP) feature extraction module; integrating the Multi-dimensional Collaborative Attention (MCA) mechanism between the neck and head networks to enhance focus on fire-related regions; adopting the Minimum Point Distance Intersection over Union (MPDIoU) loss function to optimize bounding box regression during training; and implementing selective channel pruning tailored to the modified network architecture. …”
Get full text

Sec-CLOCs: Multimodal Back-End Fusion-Based Object Detection Algorithm in Snowy Scenes by Rui Gong, Xiangsuo Fan, Dengsheng Cai, You Lu

“…Additionally, the DyHead detection head and Wise-IOU loss function are introduced to optimize YOLOv8s and improve 2D detection performance.The LIDROR algorithm preprocesses point cloud data for the SECOND detector, yielding 3D object detection results. …”
Get full text

CFD investigation and ANN prediction of heat transfer coefficient for fully developed turbulent air flow around double V-baffle turbulators by Abdulaziz Alasiri, H.E. Fawaz

“…A total of 11 hidden layers consisting of 24 neurons each has been used in constructing the ANN, and the training process is optimized using the ADAM algorithm to minimize the loss function. …”
Get full text

G-KAN: Graph Kolmogorov-Arnold Network for Node Classification Using Contrastive Learning by Lining Yuan

“…Finally, the softmax function and cross-entropy loss are employed to achieve end-to-end node classification. …”
Get full text

A novel pulse-current waveform circuit for low-energy consumption and low-noise transcranial magnetic stimulation by Xinhua Tan, Ao Guo, Jiasheng Tian, Yingwei Li, Yingwei Li, Jian Shi

“…The duration and slope of the rising and falling segments of the current waveform were adjusted by the PWM pulse duration.ResultsThe optimized current waveform, represented by three segmented functions, reduces heat loss and noise while inducing a greater change in neural membrane potential compared with those obtained with conventional symmetric waveforms. …”
Get full text

Detection method of small size defects on pipeline weld surface based on improved YOLOv7. by Xiangqian Xu, Wenting Hou, Xing Li

“…Then, the convolution module in the backbone network and the feature fusion network is replaced by the depthwise separable convolution with less computational overhead, so as to effectively reduce the network calculation, parameter quantity and model volume. Finally, the loss function CIoU of YOLOv7 is optimized to EIoU loss function to accelerate the convergence speed of the model. …”
Get full text

A techno‐economic model for enhancing PV hosting capacity through capabilities of smart inverters and soft open points by Rezvaneh Golnazari, Saeed Hasanzadeh, Ehsan Heydarian‐Forushani, Hamid Reza Sezavar, Navid Fahimi

“…Abstract The hosting capacity of solar photovoltaics (PVs) in the distribution network can be impacted by the violation of power flow constraints such as power loss, voltage, and line loadings. In this paper, a planning structure based on optimal PV placement and optimal operation of the smart inverters and soft open point (SOP) is proposed in order to investigate the loss of power, voltage profiles, line loadings and total operation cost of the distribution system. …”
Get full text

Siamese Neural Networks in Unmanned Aerial Vehicle Target Tracking Process by Athraa Sabeeh Hasan Allak, Jianjun Yi, Haider M. Al-Sabbagh, Liwei Chen

“…The results showed that the improved YOLOv5 model combined with the optimized loss function had the highest average accuracy of 47.84% and a frame rate of 28.34fps, which was better than the traditional YOLOv5 model. …”
Get full text

An improved lightweight tiny-person detection network based on YOLOv8: IYFVMNet by Fan Yang, Lihu Pan, Hongyan Cui, Linliang Zhang

“…This operation also reduces the computational cost by decreasing the amount of required feature map channels, while maintaining the effectiveness of the feature representation. (3) he Minimum Point Distance Intersection over Union loss function is employed to optimize bounding box detection during model training. (4) to construct the overall network structure, the Layer-wise Adaptive Momentum Pruning algorithm is used for thinning.ResultsExperiments on the TinyPerson dataset demonstrate that IYFVMNet achieves a 46.3% precision, 30% recall, 29.3% mAP50, and 11.8% mAP50-95.DiscussionThe model exhibits higher performance in terms of accuracy and efficiency when compared to other benchmark models, which demonstrates the effectiveness of the improved algorithm (e.g., YOLO-SGF, Guo-Net, TRC-YOLO) in small-object detection and provides a reference for future research.…”
Get full text

Switching Power Amplifier and Its Improved Current Control Strategy of Active Magnetic Bearing by HE Xi, HE Yaping, LI Jia

“…In order to optimize the performance of the magnetic bearing system, a five-axis common-bridge arm topology structure is proposed by analyzing the performance characteristics of magnetic bearing switching power amplifier, which has the advantages of less devices and less loss. …”
Get full text

Preservation effect of plant essential oil-KGM composite coating materials on the tomatoes. by Jianming Sun, Linlin Li, Xiao Pan, Yini Yao, Tiantian Wang, Hui Liu, Lei Liu, Zhao Li, Xiaofang Wang

“…As a result, cinnamon essential oil with good preservation effect was selected and combined with KGM to prepare coating materials for the preservation of tomatoes. The optimized preservation condition for tomatoes was in a solution with 0.70 g/L cinnamon essential oil, 8.20 g/L KGM and soaking time for 3.2 min, which had a good preservation effect on tomatoes with 2.5% weight loss ratio and 1.3% hardness after 10 days at room temperature, reduced the consumption of total soluble solid (TSS), delayed the decline of vitamin C (Vc) content, inhibited the increase of malondialdehyde (MDA) content, and enhanced peroxidase (POD) activity. …”
Get full text

A Novel Network for Choroidal Segmentation Based on Enhanced Boundary Information by Wenbo Huang, Chaofan Qu, Yang Yan

“…A custom loss function, specifically designed to replace the conventional Mean Squared Error (MSE) loss, is utilized to optimize network parameters, leading to further improvements in segmentation performance. …”
Get full text

The analysis of learning investment effect for artificial intelligence English translation model based on deep neural network by Yan Zhang, Shuangshuang Lyu

“…The model innovatively integrates target future contextual information with a multimodal consistency loss function, effectively capturing the interaction between text and visual information to optimize translation performance. …”
Get full text

Identification of fresh leaves of Anji White Tea: S-YOLOv10-ASI algorithm fusing asymptotic feature pyra-mid network. by Chunhua Yang, Wenxia Yuan, Qiang Zhao, Zejun Wang, Bowu Song, Xianqiu Dong, Yuandong Xiao, Shihao Zhang, Baijuan Wang

“…Intersection over Union (IoU) is used to optimize the loss function calculation. The slice-assisted super-reasoning algorithm is integrated to improve the recognition ability of YOLOv10 network for long-distance and small-target tea. …”
Get full text

The Speaker Identification Model for Air-Ground Communication Based on a Parallel Branch Architecture by Weijun Pan, Shenhao Chen, Yidi Wang, Sheng Chen, Xuan Wang

“…The CET model incorporates three key components: the Chrono Block module, the speaker embedding extraction module, and the optimized loss function module. The Chrono Block module utilizes parallel branching architecture, Bi-LSTM, and multi-head attention mechanisms to effectively extract both global and local features, addressing the challenge of short speech. …”
Get full text