Lightweight Detection of Grape Inflorescences and Fruitlets using an Improved YOLOv8 Model
Hu Guoyu;Lin Zhe;Wang Haining;Jiang Dexuan;Globally, grape cultivation spans vast areas and achieves substantial yields, making grapes and related industries vital economic pillars for many nations. In grape production, efficient and precise management during key growth stages is essential for enhancing both yield and quality. In view of the problems that during the grape inflorescences and young fruits stage, the targets are small in size, easily obscured by branches and leaves, and highly similar in color to the background, resulting in poor recognition performance of existing detection methods in complex natural environments, which in turn restricts the application of precision spraying technology. This paper establishes a dedicated dataset for grape inflorescences and young fruits in Xinjiang and proposes an improved lightweight detection model, YOLOv8-FCD. The model incorporates a PConv-based C2f＿Faster module to reduce parameter count and computational complexity, replaces the original upsampling method with the CARAFE module to enhance feature extraction capability, and introduces the Detect＿SEAM detection head to improve recognition accuracy under occlusion and small-target conditions. Experimental results show that the YOLOv8-FCD model achieves a detection precision(P) of 93.7% and a recall(R) of 87.3%, with a mean average precision(m AP) of 94.6%. Compared to the original YOLOv8n model, P improved by 8.2%, m AP increased by 2.6%, and the model size is reduced to 85.71% of the original. This model provides effective technical support for the identification of grape inflorescences and young fruits in intelligent spraying for plant protection.
Farmland Change Detection Algorithm Based on Improved BIT
Xu Shiliang;Lai Minquan;Liu Jizhong;Jiangxi Natural Resources Development Center;Farmland non-agriculturalization is a serious threat to global food security and ecological stability. Remote sensing change detection technology has become a core tool for identifying the process of farmland non-agriculturalization by virtue of its advantage of large-scale dynamic monitoring. However, existing methods face challenges in balancing the extraction of fine edge details in fragmented farmland with the maintenance of global semantic consistency in large-scale fields, often resulting in blurred edges and lost local features. To address these issues, a farmland change detection algorithm based on improved BIT, named Far-CDNet, is proposed. Firstly, a detail enhancement convolution module that connects ordinary convolution and multiple differential convolutions in parallel is introduced, and the edge detail representation capability of the feature extraction network is enhanced through dynamic weighting and residual connection. Secondly, the ordinary convolution of the semantic tokenizer in the BIT module is replaced by a deep separable convolution to enhance the local feature capture ability and generate output features with higher-level semantics, so as to improve the overall feature expression ability of the model. Finally, a residual branch is added to further integrate the local and global information before and after the Transformer. The experimental results show that the improved model F1 score is 79.18%, and IoU is 69.32%. Compared with the BIT model, the F1 score is increased by 4.17%, and IoU is increased by 4.24%.
Fine-Grained Long and Short-Term Preference Sequential Recommendation with Contrastive Learning
Yang Xingyao;Wu Yanfu;Zhang Zulian;Yu Jiong;Zhong Zhiqiang;Chen Yu;Sequence recommendation aims at item recommendation using users' long and short-term preferences, but most sequence recommendation systems face problems such as insufficient learning power and inadequate fusion of long and shortterm preferences. Aiming at the above problems, this paper proposes a fine-grained long and short-term preference sequence recommendation method based on contrastive learning. 1) To address the problem of insufficient long and short-term preference fusion, this paper proposes a long and short-term preference learning layer and a long and short-term preference fusion layer. Firstly, it splits the user behaviour sequence into multi-period sessions and extracts the user's short-term preference in each session by using gated recurrent units, and then fuses the short-term preference sequences to capture the user's long-term preference through the multi-head attention mechanism. Finally, the long-term and short-term preferences are fused adaptively based on the time span to obtain a more representative and comprehensive preference representation. 2) Aiming at the problem of insufficient learning power due to data sparsity, a preference representation comparison learning task is designed to introduce agent user preferences for comparison learning to achieve more accurate preference recommendation. The experimental results show that: compared to the sub-optimal methods, the model improves the Hit@20 metric by 9.84%, 6.40%, and 1.52%, and the MAP@20 metric by 22.64%, 2.42%, and 6.42% on three public datasets, respectively, demonstrating the effectiveness of the proposed method.
A Dynamic Prediction Model for NOx Concentration Based on MEVMD and GA-CNN+LSTM
Zhang Qifan;Hu Lina;Zeng Hao;Liu Wei;Yang Can;Coal-fired power plants serve as primary sources of NOx emissions, and the efficient operation of SCR denitrification systems is crucial for reducing pollutant emissions. However, the highly dynamic changes of data during NOx prediction processes limit the accuracy of predictive models. Therefore, a hybrid prediction framework based on modal energy difference and sample entropy, which combining variational mode decomposition(MEVMD) with genetic algorithm(GA) to optimize convolutional neural network(CNN) and long short-term memory network(LSTM), is proposed. Firstly, abnormal data are corrected using the 3σ criterion; 20 key auxiliary variables are selected via Pearson correlation coefficients. The maximum information coefficient(MIC) is employed to determine the delay time for each variable, achieving temporal alignment between features and target variables. Secondly, adaptive variational mode decomposition(VMD) precisely extracts multifrequency features from NOx time-series signals. Hyperparameters are optimized via GA to achieve adaptive modeling of multiple sub-modes. Finally, prediction results are generated through data reconstruction. Experimental results demonstrate that the proposed model achieves RMSE of 0.949 2, MAE of 0.496 9, and R2 of 0.976 7, outperforming comparison models.
The Asymptotic Limit of the Riemann Solution for the Macroscopic Production Model with Anti-Chaplygin Gas
He Weihua;Guo Lihui;This paper mainly studies the limit behavior of Riemann solutions for the macroscopic production model with antiChaplygin gas. Firstly, we investigate the Riemann problem associated with this model. Three types of Riemann solutions are obtained: a combination of contact discontinuity and rarefaction wave( J1 + R2), a combination of contact discontinuity and shock wave( J1 + S2), and Dirac shock wave( δS). Secondly, the pressure vanishing limit of the macroscopic production model of the anti-Chaplygin gas is studied. As the perturbation parameter ε decreases to the parameter ε0, which is dependent only on the initial data, it is proved that the Riemann solution(J1 + S2) converges to the δS of the anti-Chaplygin gas state equation. Moreover, when ε eventually approaches 0, the δS converges to the δS of the transport equation. Additionally, it is proved that the Riemann solution(J1 + R2) converges to the vacuum solution of the transport equation. Finally, we present some representative numerical experimental results.
Tracking the information about your manuscript
Communicate with the editorial office
Query manuscript payment status Edit officeCollecting, editing, reviewing and other affairs offices
Managing manuscripts
Managing author information and external review Expert Information Expert officeOnline Review
Online Communication with the Editorial Department
