Query：
学者姓名：邹腾跃
Refining：
Year
Type
Indexed by
Source
Complex
Co-Author
Language
Clean All
Abstract ：
BACKGROUNDThe flavor profile and product quality of white tea, heavily dependent on its place of origin, significantly influence consumers' purchasing decisions. Quantitative adulteration testing for tea origin has encountered challenges due to the poor performance in random external validation, which has severely hindered the practical application of near-infrared (NIR) technology.RESULTSThis study employs a two-dimensional convolutional neural network (2D-CNN) deep learning model combined with Gramian angular field (GAF) image coding technology (GAF-2D-CNN) to quantitatively detect geographical origin adulteration of white tea using near-infrared spectral (NIRS) data. The results demonstrate that the GAF-2D-CNN model can effectively process raw spectral data and predict the untrained random adulteration ratio data with high accuracy. The average R2 and root mean square error in the external verification of the original data reach 0.9754 and 0.0349, respectively, which meet practical production needs. Moreover, the GAF-2D-CNN significantly outperforms traditional regression models and 1D-CNN models.CONCLUSIONThis study introduces the application of the NIR spectral image coding method in tea regression and highlights the advantages of deep learning image processing in the tea industry. (c) 2025 Society of Chemical Industry.
Keyword ：
geographical adulteration geographical adulteration Gramian angular field Gramian angular field near-infrared spectral image coding near-infrared spectral image coding two-dimensional convolutional neural networks two-dimensional convolutional neural networks
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Zhang, Peng , Cheng, Jun , Chen, Qinglan et al. Near-infrared spectroscopy coupled with Gramian angular field two-dimensional convolutional neural network for white tea adulteration detection [J]. | JOURNAL OF THE SCIENCE OF FOOD AND AGRICULTURE , 2025 , 105 (11) : 6269-6279 . |
| MLA | Zhang, Peng et al. "Near-infrared spectroscopy coupled with Gramian angular field two-dimensional convolutional neural network for white tea adulteration detection" . | JOURNAL OF THE SCIENCE OF FOOD AND AGRICULTURE 105 . 11 (2025) : 6269-6279 . |
| APA | Zhang, Peng , Cheng, Jun , Chen, Qinglan , Zheng, Zhiqiang , Wei, Chengjiang , Zou, Tengyue et al. Near-infrared spectroscopy coupled with Gramian angular field two-dimensional convolutional neural network for white tea adulteration detection . | JOURNAL OF THE SCIENCE OF FOOD AND AGRICULTURE , 2025 , 105 (11) , 6269-6279 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
固定步长的自适应滤波降噪算法(LMS)的步长选择直接影响滤波性能。为此，文章提出了一种基于反正切函数的变步长LMS算法。该算法通过引入指数函数并对反正切函数进行平移变换，使得步长在误差趋于零时自适应减小，从而降低稳态误差。在MATLAB仿真环境下，分析了不同变步长LMS算法的步长变化曲线及均方误差(MSE)。仿真结果表明，相较于传统的Sigmoid变步长、箕舌线变步长及反正切的变步长方法，所提出的改进反正切变步长LMS算法具备更快的收敛速度和更高的滤波精度。
Keyword ：
反正切函数 反正切函数 变步长LMS算法 变步长LMS算法 均方误差 均方误差 指数函数 指数函数 收敛速度 收敛速度 自适应滤波 自适应滤波
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 李典航 , 王申营 , 程骏 et al. 基于改进反正切函数的变步长LMS算法研究 [J]. | 信息技术与信息化 , 2025 , 5 (04) : 118-122 . |
| MLA | 李典航 et al. "基于改进反正切函数的变步长LMS算法研究" . | 信息技术与信息化 5 . 04 (2025) : 118-122 . |
| APA | 李典航 , 王申营 , 程骏 , 彭云锋 , 邹腾跃 . 基于改进反正切函数的变步长LMS算法研究 . | 信息技术与信息化 , 2025 , 5 (04) , 118-122 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
当面对具有混响和强噪声的复杂环境时，传统的广义二次互相关时延估计算法往往存在精度不足的问题。为此，提出了一种改进的自适应时延估计算法。该算法的核心思想是利用前置自适应滤波器对声源信号进行降噪处理并获取信噪比参数，将信噪比参数与Sigmoid函数相结合并引入相位变换(PHAT)加权函数的加权因子中，使得时延估计算法具备根据噪声强度自适应调整加权的能力。以人声声源信号为例，利用Matlab进行不同信噪比环境下的性能仿真试验，并在五元十字麦克风阵列下完成声源定位的仿真。试验结果表明，在不同低信噪比下，改进算法的声源定位距离偏差比优于传统算法。因此，在复杂环境中，改进算法具有更高的定位精度和鲁棒性。该研究为复杂环境下的声源定位提供了新的解决方案，对于提升智能会议系统、远程监控系统、声源探测仪表等领域的声源定位精度具有重要意义。
Keyword ：
Sigmoid函数 Sigmoid函数 信噪比 信噪比 加权因子 加权因子 声源定位 声源定位 时延估计 时延估计 相位变换 相位变换 自适应 自适应
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 李典航 , 王申营 , 程骏 et al. 结合Sigmoid函数的自适应时延估计算法研究 [J]. | 自动化仪表 , 2025 , 46 (04) : 31-35,41 . |
| MLA | 李典航 et al. "结合Sigmoid函数的自适应时延估计算法研究" . | 自动化仪表 46 . 04 (2025) : 31-35,41 . |
| APA | 李典航 , 王申营 , 程骏 , 彭云锋 , 邹腾跃 . 结合Sigmoid函数的自适应时延估计算法研究 . | 自动化仪表 , 2025 , 46 (04) , 31-35,41 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Strawberries are susceptible to various diseases during their growth, and leaves may show signs of diseases as a response. Given that these diseases generate yield loss and compromise the quality of strawberries, timely detection is imperative. To automatically identify diseases in strawberry leaves, a KTD-YOLOv8 model is introduced to enhance both accuracy and speed. The KernelWarehouse convolution is employed to replace the traditional component in the backbone of the YOLOv8 to reduce the computational complexity. In addition, the Triplet Attention mechanism is added to fully extract and fuse multi-scale features. Furthermore, a parameter-sharing diverse branch block (DBB) sharing head is constructed to improve the model's target processing ability at different spatial scales and increase its accuracy without adding too much calculation. The experimental results show that, compared with the original YOLOv8, the proposed KTD-YOLOv8 increases the average accuracy by 2.8% and reduces the floating-point calculation by 38.5%. It provides a new option to guide the intelligent plant monitoring system and precision pesticide spraying system during the growth of strawberry plants.
Keyword ：
deep learning deep learning smart agriculture smart agriculture strawberry disease strawberry disease target detection target detection
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | He, Yuelong , Peng, Yunfeng , Wei, Chuyong et al. Automatic Disease Detection from Strawberry Leaf Based on Improved YOLOv8 [J]. | PLANTS-BASEL , 2024 , 13 (18) . |
| MLA | He, Yuelong et al. "Automatic Disease Detection from Strawberry Leaf Based on Improved YOLOv8" . | PLANTS-BASEL 13 . 18 (2024) . |
| APA | He, Yuelong , Peng, Yunfeng , Wei, Chuyong , Zheng, Yuda , Yang, Changcai , Zou, Tengyue . Automatic Disease Detection from Strawberry Leaf Based on Improved YOLOv8 . | PLANTS-BASEL , 2024 , 13 (18) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Background: Sleep apnea, including obstructive and central types, is a common sleep disorder. Traditional polysomnography is inconvenient and relies on professional interpretation. Advances in artificial intelligence and large-scale clinical databases offer more efficient and cost-effective detection methods. In the current study, bidirectional long short-term memory (Bi-LSTM), as a powerful tool from the recurrent neural network, is used to achieve sleep apnea incident detection with only the input of an electroencephalogram (EEG) signal.Methods: The MIT-BIH polysomnography database is utilized. The Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) algorithm denoises the signals. Features are extracted from the denoised EEG segments on a one-second basis, including energy values of four bands, sample entropy, and sequence complexity. Sleep state calibration is performed over thirty-second periods, grouping all sleep apnea and hypopnea events. The Whale Optimization Algorithm (WOA) optimizes the network's hyperparameters. The Bi-LSTM model is used for event detection, with K-fold cross-validation to evaluate performance.Results: Compared to other commonly used algorithms, the Bi-LSTM model proposed in the current study achieves up to 96.7% of accuracy, 95.3% of sensitivity, 97.9% of specificity, and 96.4% of F1 score. The best result occurs when the K value is 7.Conclusion: An EEG-based Bidirectional Recurrent Neural Network (EBRNN) is introduced for sleep apnea detection, demonstrating its simplicity and superior performance. Compared to polysomnography, the Bi-LSTM model is not only accurate but also significantly simplifies the detection process, offering a powerful tool and viable strategy for sleep apnea detection. © 2024, The Authors. All rights reserved.
Keyword ：
Empirical mode decomposition Empirical mode decomposition Image segmentation Image segmentation Long short-term memory Long short-term memory Signal denoising Signal denoising Sleep research Sleep research
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wang, Shenying , Huang, Xuanyu , Cheng, Jun et al. Eeg-Based Bidirectional Recurrent Neural Network in Sleep Apnea Incident Detection [J]. | SSRN , 2024 . |
| MLA | Wang, Shenying et al. "Eeg-Based Bidirectional Recurrent Neural Network in Sleep Apnea Incident Detection" . | SSRN (2024) . |
| APA | Wang, Shenying , Huang, Xuanyu , Cheng, Jun , Xin, Jiawei , Zou, Tengyue . Eeg-Based Bidirectional Recurrent Neural Network in Sleep Apnea Incident Detection . | SSRN , 2024 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The traditional laser SLAM (Simultaneous Localization and Mapping) algorithm uses the global relative poses and local ones to form residual blocks. Its constructed map is not smooth enough and the constraint construction is too simplex under some special scenarios. Thus, this paper proposes an odometer constraint fusion method called FOSLAM (Fusion Odometer SLAM) to construct residual blocks between constrains and solve the nonlinear least squares by Ceres. The effectiveness and accuracy of this method have been verified through comparative experiments. Experimental results showed that without increasing the time and space complexity, by involving the odometer constraint into the SLAM optimization process, the convergence of scan matching scores can be improved and the constructed grid map edges are smoother and the jagged phenomenon can be reduced. Under sophisticated scene, FOSLAM is able to acquire more accurate maps and laser odometer trajectory than Cartographer method. Therefore, it is suitable to be used on indoor robot for cleaning and inspection and can be further deployed on autonomous unmanned vehicles involving spatial visualization and neuro-heuristic guidance.
Keyword ：
Back-end optimization Back-end optimization Ceres Ceres Laser SLAM Laser SLAM Odometer constraint fusion Odometer constraint fusion Redisual blocks Redisual blocks
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Huang, Haojun , Yang, Puxian , Cai, Shengqing et al. A Construction Optimization for Laser SLAM Based on Odometer Constraint Fusion [J]. | INFORMATION TECHNOLOGY AND CONTROL , 2024 , 53 (2) . |
| MLA | Huang, Haojun et al. "A Construction Optimization for Laser SLAM Based on Odometer Constraint Fusion" . | INFORMATION TECHNOLOGY AND CONTROL 53 . 2 (2024) . |
| APA | Huang, Haojun , Yang, Puxian , Cai, Shengqing , Li, Jixiang , Zheng, Yuda , Zou, Tengyue . A Construction Optimization for Laser SLAM Based on Odometer Constraint Fusion . | INFORMATION TECHNOLOGY AND CONTROL , 2024 , 53 (2) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
针对现有检测模型不能满足在自然环境中准确识别多种类柑橘病虫害的问题，提出一种基于改进YOLOv5s模型的常见柑橘病虫害检测方法。改进模型引入ConvNeXtV2模型，构建一个CXV2模块替换YOLOv5s的C3模块，增强提取特征的多样性；添加了动态检测头DYHEAD，提高模型对不同空间尺度、不同任务目标的处理能力；采用CARAFE上采样模块，提高特征提取效率。结果显示，改进后的YOLOv5s-CDC的召回率和平均精度均值分别为81.6%、87.3%，比原模型分别提高了4.9、3.4百分点。与其他YOLO系列模型在多个场景下的检测对比，具有更高的准确率和较强的鲁棒性。结果表明，该方法可用于自然复杂环境下的柑橘病虫害的检测。
Keyword ：
YOLOv5s YOLOv5s 深度学习 深度学习 病虫害 病虫害 目标检测 目标检测
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 郑宇达 , 陈仁凡 , 杨长才 et al. 基于改进YOLOv5s模型的柑橘病虫害识别方法 [J]. | 华中农业大学学报 , 2024 , 43 (02) : 134-143 . |
| MLA | 郑宇达 et al. "基于改进YOLOv5s模型的柑橘病虫害识别方法" . | 华中农业大学学报 43 . 02 (2024) : 134-143 . |
| APA | 郑宇达 , 陈仁凡 , 杨长才 , 邹腾跃 . 基于改进YOLOv5s模型的柑橘病虫害识别方法 . | 华中农业大学学报 , 2024 , 43 (02) , 134-143 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wang, Shenying , Zou, Tengyue , Huang, Xuanyu et al. EEG-based bidirectional recurrent neural network in sleep apnoea incident detection [J]. | JOURNAL OF SLEEP RESEARCH , 2024 , 33 . |
| MLA | Wang, Shenying et al. "EEG-based bidirectional recurrent neural network in sleep apnoea incident detection" . | JOURNAL OF SLEEP RESEARCH 33 (2024) . |
| APA | Wang, Shenying , Zou, Tengyue , Huang, Xuanyu , Cheng, Jun , Pan, Xiaodong , Xin, Jiawei . EEG-based bidirectional recurrent neural network in sleep apnoea incident detection . | JOURNAL OF SLEEP RESEARCH , 2024 , 33 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
【目的】解决自然环境下不同成熟度桃子快速准确检测的问题，课题组提出一种基于改进YOLOv5s的目标检测算法YOLO-Faster。【方法】使用YOLOv5s网络模型作为基础网络，将主干特征提取网络替换为FasterNet，使模型轻量化，并在主干和颈部之间增加串联的CBAM卷积注意力模块和常规卷积块，增强对图像重要特征的捕捉与表达，同时引入SIoU损失函数缓解预测框与真实框之间方向的不匹配。【结果】改进后模型的m AP为88.6%，与YOLOv5s相比提升1个百分点，模型权重缩减39.4%，浮点运算量降低44.3%，在GPU、CPU上的单张图像平均检测时间分别减少12.6%和24%。此外，本研究将训练好的模型部署到嵌入式设备Jetson Nano上，模型在Jetson Nano上的检测时间比YOLOv5s减少30.4%。【结论】改进后的轻量级模型能够快速准确地检测自然环境下不同成熟度的桃子，可以为桃子采摘机器人的视觉识别系统提供技术支持。
Keyword ：
FasterNet FasterNet YOLOv5s YOLOv5s 快速识别 快速识别 桃子成熟度 桃子成熟度 注意力机制 注意力机制 目标检测 目标检测
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 曾俊 , 陈仁凡 , 邹腾跃 . 基于改进YOLO的自然环境下桃子成熟度快速检测模型 [J]. | 南方农机 , 2023 , 54 (24) : 24-27,41 . |
| MLA | 曾俊 et al. "基于改进YOLO的自然环境下桃子成熟度快速检测模型" . | 南方农机 54 . 24 (2023) : 24-27,41 . |
| APA | 曾俊 , 陈仁凡 , 邹腾跃 . 基于改进YOLO的自然环境下桃子成熟度快速检测模型 . | 南方农机 , 2023 , 54 (24) , 24-27,41 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Micro-groove heat pipe is a good choice for heat dissipation of large CNC lathes. Its thermal conductivity is limited by a variety of heat transfer limits, which mainly depend on the geometrical parameters, the status of working fluid, the structure of the wick and the working temperature. Through mathematical modeling the optimized geometric parameters are chosen as 9.38 mm inner hole diameter and 55 grooves with 0.32 mm top side width, 0.31 mm groove depth, 0.31458 mm bottom width. Analysis of gas-liquid two-phase flow is taken to find out that the best filling rate is 22.62% for the working fluid. An orthogonal experiment is also carried out to get the suitable length, working temperature and inclination angle for the micro-groove heat pipe. From the experimental results, the heating temperature is found to have the largest impact on the heat transfer rate, followed by the heat pipe length, and the inclination angle has the least impact. Moreover, 50°C bath temperature, 160 mm heat pipe length and 90° inclination angle are the most suitable environmental parameters for the flat micro-groove heat pipe under the experimental circumstance. © 2023, Politechnica University of Bucharest. All rights reserved.
Keyword ：
Geometry Geometry Heat pipes Heat pipes Heat transfer Heat transfer Two phase flow Two phase flow
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Li, Xibing , Li, Weixiang , Li, Jixiang et al. PARAMETER OPTIMIZATION FOR FLAT MICRO-GROOVE HEAT PIPE BY GAS-LIQUID TWO-PHASE FLOW ANALYSIS [J]. | UPB Scientific Bulletin, Series D: Mechanical Engineering , 2023 , 85 (2) : 187-198 . |
| MLA | Li, Xibing et al. "PARAMETER OPTIMIZATION FOR FLAT MICRO-GROOVE HEAT PIPE BY GAS-LIQUID TWO-PHASE FLOW ANALYSIS" . | UPB Scientific Bulletin, Series D: Mechanical Engineering 85 . 2 (2023) : 187-198 . |
| APA | Li, Xibing , Li, Weixiang , Li, Jixiang , Huang, Liduan , Zou, Tengyue . PARAMETER OPTIMIZATION FOR FLAT MICRO-GROOVE HEAT PIPE BY GAS-LIQUID TWO-PHASE FLOW ANALYSIS . | UPB Scientific Bulletin, Series D: Mechanical Engineering , 2023 , 85 (2) , 187-198 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
