检索结果-南通市图书馆

A novel approach based on a modified mask R-CNN for the weight prediction of live pigs

在线全文

学校读者我要写书评

暂无评论

Artificial Intelligence in Agriculture 2024年第2期12卷 19-28页

作者： Chuanqi Xie Yuji Cang Xizhong Lou hua Xiao Xing Xu Xiangjun li Weidong Zhou State Key Laboratory for Managing Biotic and chemical Threats to the Quality and Safety of Agro-Products Institute of Animal Husbandry and Veterinary ScienceZhejiang Academy of Agricultural SciencesHangzhou 310021China College of information Engineering China Jiliang UniversityHangzhou 310018China College of Hydraulic Engineering Zhejiang Tongji Vocational College of Science and TechnologyHangzhou 311231China

Since determining the weight of pigs during large-scale breeding and production is challenging,using non-contact estimation methods is *** study proposed a novel pig weight prediction method based on a mod-ified mask region-convolutional neural network(mask R-CNN).The modified approach used ResNeSt as the backbone feature extraction network to enhance the image feature extraction *** feature pyramid net-work(FPN)was added to the backbone feature extraction network for multi-scale feature *** channel at-tention mechanism(CAM)and spatial attention mechanism(SAM)were introduced in the region proposal network(RPN)for the adaptive integration of local features and their global dependencies to capture global in-formation,ultimately improving image segmentation *** modified network obtained a precision rate(P),recall rate(R),and mean average precision(MAP)of 90.33%,89.85%,and 95.21%,respectively,effectively segmenting the pig regions in the *** image features,namely the back area,body length,body width,average depth,and eccentricity,were *** pig depth images were used to build five regression algo-rithms(ordinary least squares(OLS),AdaBoost,CatBoost,XGBoost,and random forest(RF))for weight value *** achieved the best prediction result with a coefficient of determination(R^(2))of 0.987,a mean absolute error(MAE)of 2.96 kg,a mean square error(MSE)of 12.87 kg^(2),and a mean absolute percentage error(MAPE)of 8.45%.The results demonstrated that the machine learning models effectively predicted the weight values of the pigs,providing technical support for intelligent pig farm management.

关键词： Deep learning Modified mask R-CNN Image processing Pig weight Prediction

Co-incorporation of Chinese milk vetch(Astragalus sinicus L.)and chemical fertilizers alters microbial functional genes supporting short-time scale positive nitrogen priming effects in paddy soils

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Pedosphere 2024年第3期34卷 567-576页

作者： limin WANG Chunmei HE Dongfeng huaNG Juhua YU Cailing liU Qinghua li Yibin huaNG Juan li Fei WANG Institute of Resources Environment and Soil FertilizerSoil and Fertilizer InstituteFujian Academy of Agricultural SciencesFuzhou 350012(China) School of Resources and Environment Engineering Anhui UniversityHefei 230601(China)

Nitrogen(N)priming is a microbially mediated biochemical process as affected by different incorporation ***,little information is known about the microbial mechanisms driving the response of N priming to co-operation of Chinese milk vetch(CMV,Astragalus sinicus L.)and different rates of chemical fertilizers in paddy soils in South ***,an anaerobic incubation experiment was conducted to study N priming effects(PE)and their relationships with soil microbial functional genes after CMV incorporation alone(M),co-incorporation of CMV with 100%(normal dosage)chemical fertilizers(MC100),and co-incorporation of CMV with 80% chemical fertilizers(MC80).Co-incorporation of CMV and chemical fertilizers enhanced the short-time scale(the first 20 d of incubation)positive PE of N,while no significant differences existed among the three treatments on day 60 or 90 of incubation(P>0.05).Compared with the M treatment,gross priming effect(GPE)in the MC100 and MC80 treatments significantly increased by 34.0% and 31.3%,respectively,and net priming effect(NPE)increased by 47.7% and 47.8%,respectively,during the first 20 d of incubation(Plikely attributed to soil nutrient availability and added substrate *** MC100 and MC80 treatments increased the gdhA gene abundance by 5.0% and 9.8%,increased the gdh2 gene abundance by 12.7% and 45.7%,and increased the nasB gene abundance by 9.5% and 41.4%,respectively,in comparison with the M treatment on day 20 of *** analyses indicated that soil microbial functional genes involved in N mineralization(gdhA and gdh2),assimilatory nitrate reduction(nasB),and nitrification(amoB)were significantly correlated with N priming under different incorporation practices during the incubation period(Pchemical fertilizers can regulate soil microbial community functional gene structure,which may accelerate mineralization and assimilatory nitrate reduction and inhibit

关键词： green manure gross priming effect incorporation practice microbial mechanism net priming effect rice field soil

A review on plasma-based CO_(2) utilization:process considerations in the development of sustainable chemical production

在线全文

学校读者我要写书评

暂无评论

Plasma Science and Technology 2024年第9期26卷 1-16页

作者： Sirui li Giulia De FEliCE Simona EICHKORN Tao SHAO Fausto GALLUCCI Department of chemical Engineering and Chemistry Eindhoven University of TechnologyEindhoven 5612 AZthe Netherlands Institute of Electrical Engineering Chinese Academy of SciencesBeijing 100190People’s Republic of China

Plasma-based processes,particularly in carbon capture and utilization,hold great potential for addressing environmental challenges and advancing a circular carbon *** significant progress has been made in understanding plasma-induced reactions,plasma-catalyst interactions,and reactor development to enhance energy efficiency and conversion,there remains a notable gap in research concerning overall process *** review emphasizes the critical need for considerations at the process level,including integration and intensification,to facilitate the industrialization of plasma technology for chemical *** centered on the development of plasma-based processes are made with a primary focus on CO_(2) conversion,offering insights to guide future work for the transition of the technology from laboratory scale to industrial *** of current research gaps,especially in upscaling and integrating plasma reactors with other process units,is the key to addressing critical *** review further delves into relevant research in process evaluation and assessment,providing methodological insights and highlighting key factors for comprehensive economic and sustainability ***,recent advancements in novel plasma systems are reviewed,presenting unique advantages and innovative concepts that could reshape the future of process *** review provides essential information for navigating the path forward,ensuring a comprehensive understanding of challenges and opportunities in the development of plasma-based CCU process.

关键词： non-thermal plasma carbon capture and utilization process integration process intensification techno-economic analysis life cycle analysis

2001-2020年山东省花生生产地域格局变化及影响因素

在线全文

学校读者我要写书评

暂无评论

中国油料作物学报 2024年第4期46卷 737-746页

作者：张晓艳殷芳杨丽萍王兆华山东省农业科学院农业信息与经济研究所山东济南250100

基于2001-2020年统计数据,运用数理统计和区域差异法,分析了山东省花生生产地域格局的变化特征及其影响因素,为山东省花生生产的可持续发展提供科学决策依据。研究表明,近20年来,山东省花生生产呈“两减一增”局面,面积和产量分别下降33... 详细信息

基于2001-2020年统计数据,运用数理统计和区域差异法,分析了山东省花生生产地域格局的变化特征及其影响因素,为山东省花生生产的可持续发展提供科学决策依据。研究表明,近20年来,山东省花生生产呈“两减一增”局面,面积和产量分别下降33.00%和22.34%、单产增加15.93%;花生生产区域向以临沂为中心的鲁西南地区和烟台、威海、青岛所处的胶东东部地区集中;20年中有4年花生面积贡献率大于单产贡献率,有16年小于单产贡献率。研究发现,鲁东和鲁西南地区经济作物面积占比减少,是导致山东省花生种植面积下降的直接影响因素;山东花生净利润和成本利润率降低,特别是近年来的“断崖式”下跌,是导致花生种植面积和总产量逐年下降的根本原因;国内外贸易市场不稳定带来的风险,也制约了花生产业的发展。

关键词：山东省花生生产特征地域格局变化影响因素

Spatial and Temporal Variability of N, P and K Balances for Agroecosystems in china

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

Pedosphere 2005年第3期15卷 347-355页

作者： SHENRun-Ping SUNBo ZHAOQi-Guo InstituteofSoilScience ChineseAcademyofSciencesNanjing210008（China）／／DepartmentofSpatialInformationScienceNanjingUniversityofInformationScienceandTechnologyNanjing210044（China） InstituteofSoilScience ChineseAcademyofSciencesNanjing210008（China）

Nitrogen, phosphorus, and potassium balances for agroecosystems in china from 1993 to 2001 were calculated at national and provincial levels using statistical data and related parameters, and their spatial and temporal variabilities were analyzed with GIS to estimate the potential impacts of nutrient N, P and K surpluses or deficits to soil, water and air. At the national scale, the N and P balances from 1993 to 2001 showed a surplus, with the nitrogen surplus remaining relatively stable from 1997—2001. Although during this period the P surplus pattern was similar to N, it had smaller values and kept increasing as the use of phosphate fertilizer increased year by year. However, K was deficient from 1993 to 2001 even though from 1999 to 2001 the K deficit decreased. The spatial analysis revealed higher N surpluses in the more developed southeastern provinces and lowest in the western and northern provinces where there was less chemical fertilizer input. The serious K deficit mainly occurred in Shanghai and Beijing municipalities, Jiangsu, Zhejiang and Hubei provinces, and Xinjiang autonomous regions. For the years 1992, 1996 and 2001, N surpluses and K deficits had significant positive spatial correlations with per capita gross domestic product (GDP), per capita gross industrial output value, and per capita net income of rural households. This showed that the level of economic development played an important role on nutrient balances in the agroecosystems.

关键词： china economic factors NPK balance spatial and temporal variability

Simulation Study on Translocation of Confined Chain Through Interacting Nanopore

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of chemical Physics 2012年第3期25卷 308-312,373页

作者： Zi-bin Zhou hua-yu li Yong-jun Xie Department of information Science Anhui Economic Management Institute Hefei 230059 China School of Chemistry and Materials Science University of Science and Technology of China Hefei230026 China

The translocation of a confined polymer chain through an interacting nanopore has been studied using two-dimensional bond fluctuation model with Monte Carlo dynamics. For different pore-polymer interaction, the average escaping time （Tesc） of the polymer chain through the nanopore, increases roughly linearly with the chain length and the nanopore length~ respectively. However~ the large repulsive and attractive pore-polymer interaction adds the difficulty of the monomers of the chain entering and leaving the nanopore, respectively, leading to the nonmonotonical dependence of （Tesc） on the pore-polymer interaction. The detailed translocation dynamics of the chain through the interacting nanopore is inves- tigated too.

关键词： Monte Carlo simulation Polymer translocation Pore-polymer interaction Escaping time

Performance Analysis for Multimedia Communication Systems with a Multilayer Queuing network Model

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

china Communications 2018年第8期15卷 67-76页

作者： Xuehua Tang Zhongyuan Wang Xiaojun li Zhen Han Zheng He Youming Fu School of Remote Sensing and information Engineering Wuhan UniversityWuhan 430079China Research Institute of Wuhan University in Shenzhen Shenzhen 518000China School of Computer Science Wuhan UniversityWuhan 430072China State Grid economic and Technological Research Institute CO. LTDBeijing 102209China

Software performance evaluation in multimedia communication systems is typically formulated into a multi-layered client-server queuing network(MLCSQN) problem. However, the existing analytical methods to MLCSQN model cannot provide satisfactory solution in terms of accuracy, convergence and consideration of interlocking effects. To this end, this paper proposes a heuristic solving method for MLCSQN model to boost the performance prediction of distributed multimedia software systems. The core concept of this method is referred to as the basic model, which can be further decomposed into two sub-models: client sub-model and server sub-model. The client sub-model calculates think time for server sub-model, and the server sub-model calculates waiting time for client sub-model. Using a breadthfirst traversal from leaf nodes to the root node and vice versa, the basic model is then adapted to MLCSQN, with net sub-models iteratively resolved. Similarly, the interlocking problem is effectively addressed with the help of the basic model. This analytical solver enjoys advantages of fast convergence, independence on specific average value analysis(MVA) methods and eliminating interlocking *** experimental results on accuracy and computation efficiency verify its superiority over anchors.

关键词： multimedia communication system queuing network performance evaluation

A Group Collaborative Evaluation Method Based on Web2

在线全文

学校读者我要写书评

暂无评论

A Group Collaborative Evaluation Method Based on Web2

2011 World Congress on Engineering and Technology(CET 2011)

作者： li Yu School of information Renmin University of china Beijing P. R. china Meiyun Zuo School of information Renmin University of china Beijing P. R. china Xi Jiang School of information Renmin University of china Beijing P. R. china

Evaluation is one of most important issues in management science. Enlightened by Web2.0, in this paper, the idea on collaboration is employed in management evaluation, and a new management evaluation method, Collaborative Evaluation （CE） is proposed. Its advantage, disadvantage and application processes are detailed. A special CE method is designed, and is verified in the experiment. The experiment shows there is great advantage for CE method over traditional method.

关键词： Collaborative Evaluation Evaluation Theory Evaluation Method Web2.0 Collaborative Management

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

混合卷积神经网络用于高光谱小麦品种鉴别

光谱学与光谱分析 2024年第3期44卷 807-813页

作者：李国厚李泽旭金松林赵文义潘细朋梁政秦莉张卫东河南科技学院信息工程学院河南新乡453003 北京邮电大学人工智能学院北京100876 桂林电子科技大学计算机与信息安全学院广西桂林541004 安徽大学互联网学院安徽合肥230039 宁波大学信息科学与工程学院浙江宁波315211

不同品种的小麦满足了市场的不同需求,同时也会带来小麦品种混杂的风险。为了提高小麦品种的纯度进而提高选种、育种、加工等环节的经济价值,小麦种子的鉴别起到关键作用。传统的小麦品种纯度理化分析鉴别方法,鉴定时间长且破坏种子,已... 详细信息

不同品种的小麦满足了市场的不同需求,同时也会带来小麦品种混杂的风险。为了提高小麦品种的纯度进而提高选种、育种、加工等环节的经济价值,小麦种子的鉴别起到关键作用。传统的小麦品种纯度理化分析鉴别方法,鉴定时间长且破坏种子,已不能满足现代农业的迫切需要。高光谱成像作为近年来发展迅速的一种快速、高效、无损的新型鉴别技术,在种子品种鉴别领域取得了显著成效。然而,已有的大多数高光谱分类方法仅利用光谱信息,没有充分考虑空间信息,分类效果较差。为了解决上述问题,利用高光谱成像设备采集8个品种的小麦种子正背面的高光谱图像,基于这些高光谱数据集,提出一种基于注意力机制的混合卷积神经网络的高光谱小麦品种鉴别方法,主要利用三维卷积和二维卷积的互补优势特性提取小麦的有价值特征,进而提高小麦品种的鉴别效果。具体而言,首先提取小麦品种的感兴趣区域,并利用多元散射校正方法削弱由于散射水平差异带来的同一品种的光谱差异。同时,利用主成分分析方法减少三维数据的无用光谱波段,进而保留并压缩对鉴别小麦品种有价值的特征。随后,利用三维卷积获取空间维度和不同光谱间的特征信息,二维卷积获取空间信息和图像的自身固有的特征信息,并在二维卷积模型中引入注意力机制进一步增强图像的特征信息的提取。最后在全连接层实现同一区域不同小麦品种的鉴别。实验表明,所提出的方法比其他方法具有较好的分类性能,分类准确率达97.92%。此外,所提出的方法对小样本数据也具有较好的分类性能。总的来说,提出的方法对于高光谱小麦种子鉴别具有较好的有效性和鲁棒性,为小麦种子的在线鉴别提供了一种新的方法。

关键词：高光谱成像小麦品种注意力机制混合卷积