响应面法是一种常见的多元统计优化方法,广泛应用于食品工业、工程技术、生物科学、电子信息等多个研究领域。基于中国知网和Web of Science数据库,收集国内外相关文献,采用CiteSpace软件可视化分析响应面法在国内外食用菌研究中的应用...响应面法是一种常见的多元统计优化方法,广泛应用于食品工业、工程技术、生物科学、电子信息等多个研究领域。基于中国知网和Web of Science数据库,收集国内外相关文献,采用CiteSpace软件可视化分析响应面法在国内外食用菌研究中的应用概况,并综述响应面法在食用菌各研究领域的应用成果,以期为响应面法在食用菌研究中进一步应用提供思路和参考。展开更多
This paper presents a method to reconstruct 3-D models of trees from terrestrial laser scan(TLS)point clouds.This method uses the weighted locally optimal projection(WLOP)and the AdTree method to reconstruct detailed ...This paper presents a method to reconstruct 3-D models of trees from terrestrial laser scan(TLS)point clouds.This method uses the weighted locally optimal projection(WLOP)and the AdTree method to reconstruct detailed 3-D tree models.To improve its representation accuracy,the WLOP algorithm is introduced to consolidate the point cloud.Its reconstruction accuracy is tested using a dataset of ten trees,and the one-sided Hausdorff distances between the input point clouds and the resulting 3-D models are measured.The experimental results show that the optimal projection modeling method has an average one-sided Hausdorff distance(mean)lower by 30.74%and 6.43%compared with AdTree and AdQSM methods,respectively.Furthermore,it has an average one-sided Hausdorff distance(RMS)lower by 29.95%and 12.28%compared with AdTree and AdQSM methods.Results show that the 3-D model generated fits closely to the input point cloud data and ensures a high geometrical accuracy.展开更多
文摘响应面法是一种常见的多元统计优化方法,广泛应用于食品工业、工程技术、生物科学、电子信息等多个研究领域。基于中国知网和Web of Science数据库,收集国内外相关文献,采用CiteSpace软件可视化分析响应面法在国内外食用菌研究中的应用概况,并综述响应面法在食用菌各研究领域的应用成果,以期为响应面法在食用菌研究中进一步应用提供思路和参考。
基金supported in part by the National Natural Science Foundation of China(Nos.42271343,42177387)the Fund of State Key Laboratory of Remote Sensing Information and Image Analysis Technology of Beijing Research Institute of Uranium Geology under(No.6142A010403)
文摘This paper presents a method to reconstruct 3-D models of trees from terrestrial laser scan(TLS)point clouds.This method uses the weighted locally optimal projection(WLOP)and the AdTree method to reconstruct detailed 3-D tree models.To improve its representation accuracy,the WLOP algorithm is introduced to consolidate the point cloud.Its reconstruction accuracy is tested using a dataset of ten trees,and the one-sided Hausdorff distances between the input point clouds and the resulting 3-D models are measured.The experimental results show that the optimal projection modeling method has an average one-sided Hausdorff distance(mean)lower by 30.74%and 6.43%compared with AdTree and AdQSM methods,respectively.Furthermore,it has an average one-sided Hausdorff distance(RMS)lower by 29.95%and 12.28%compared with AdTree and AdQSM methods.Results show that the 3-D model generated fits closely to the input point cloud data and ensures a high geometrical accuracy.
