[Objective] The aim was to establish a model based on spatial scene similarity, for which soil, slope, transport, water conservancy, light, social economic factors in suitable planting areas were all considered. A new...[Objective] The aim was to establish a model based on spatial scene similarity, for which soil, slope, transport, water conservancy, light, social economic factors in suitable planting areas were all considered. A new suitable planting area of flue-cured tobacco was determined by comparison and analysis, with consideration of excellent area. [Method] Totaling thirty natural factors were chosen, which were clas- sified into nine categories, from Longpeng Town （LP） and Shaochong Town （SC） in Shiping County in Honghe Hani and Yi Autonomous Prefecture. [Result] According to weights, the factors from high to low were as follows： soil〉light〉elevation〉slope〉 water conservancy〉transport〉baking facility〉planting plans over the years〉others. The similarity of geographical conditions in the area was 0.894 3, which indicated that the planting conditions in the two regions are similar. If farmer population in unit area, farmland quantity for individual farmer, labors in every household, activity in planting flue-cured tobacco and work of local instructor were considered, the weights of different factors were as follows： farmer population in unit area〉farmland quantity for individual farmer〉farmers＇ activity in planting flue-cured tobacco〉educational back- ground〉labor force in every household〉instructor〉population of farmers＇ children at- tending school. The similarity of geographical conditions was 0.703 1, which indicated that it is none-natural factors that influence yield and quality of flue-cured tobacco. [Conclusion] According to analysis on suitable planting area of flue-cured tobacco based on assessment of spatial scene similarity, similarity of growing conditions in two spatial scenes can be analyzed and evaluated, which would promote further exploration on, influencing factors and effects on tobacco production.展开更多
Rapid and accurate acquisition and analysis of information is crucial for emergency management,but traditional methods have limitations such as incomplete information acquisition and slow processing speed.The natural ...Rapid and accurate acquisition and analysis of information is crucial for emergency management,but traditional methods have limitations such as incomplete information acquisition and slow processing speed.The natural language oriented spatial scene reconstruction method provides a new solution for emergency management,but existing generative models have limited understanding of spatial relationships and lack high-quality training samples.To address these issues,this paper proposes a novel spatial scene reconstruction framework.Specifically,the BERT based spatial information knowledge graph extraction method is used to encode the input text,label and classify the encoded text,identify spatial objects and relationships in the text,and accurately extract spatial information.Additionally,a large number of manual experiments were conducted to explore quantitative biases in human spatial cognition,and based on the obtained biases,a greedy resolution method based on cost functions was used to fine tune the layout of conflicting spatial objects and solve the conflicting spatial information in the spatial information knowledge graph.Finally,use graph convolutional neural networks to obtain scene knowledge graph embeddings that consider spatial constraints.In addition,a high-quality training sample set of“text-scene-knowledge graph”was constructed.展开更多
文摘[Objective] The aim was to establish a model based on spatial scene similarity, for which soil, slope, transport, water conservancy, light, social economic factors in suitable planting areas were all considered. A new suitable planting area of flue-cured tobacco was determined by comparison and analysis, with consideration of excellent area. [Method] Totaling thirty natural factors were chosen, which were clas- sified into nine categories, from Longpeng Town （LP） and Shaochong Town （SC） in Shiping County in Honghe Hani and Yi Autonomous Prefecture. [Result] According to weights, the factors from high to low were as follows： soil〉light〉elevation〉slope〉 water conservancy〉transport〉baking facility〉planting plans over the years〉others. The similarity of geographical conditions in the area was 0.894 3, which indicated that the planting conditions in the two regions are similar. If farmer population in unit area, farmland quantity for individual farmer, labors in every household, activity in planting flue-cured tobacco and work of local instructor were considered, the weights of different factors were as follows： farmer population in unit area〉farmland quantity for individual farmer〉farmers＇ activity in planting flue-cured tobacco〉educational back- ground〉labor force in every household〉instructor〉population of farmers＇ children at- tending school. The similarity of geographical conditions was 0.703 1, which indicated that it is none-natural factors that influence yield and quality of flue-cured tobacco. [Conclusion] According to analysis on suitable planting area of flue-cured tobacco based on assessment of spatial scene similarity, similarity of growing conditions in two spatial scenes can be analyzed and evaluated, which would promote further exploration on, influencing factors and effects on tobacco production.
基金supported in part by the Fundamental Research Funds for the Central Universities of Beijing University of Chemical Technology(Grant No.BUCTRC202132)the National Natural Science Foundation of China(Grant Nos.42371476 and 41971366).
文摘Rapid and accurate acquisition and analysis of information is crucial for emergency management,but traditional methods have limitations such as incomplete information acquisition and slow processing speed.The natural language oriented spatial scene reconstruction method provides a new solution for emergency management,but existing generative models have limited understanding of spatial relationships and lack high-quality training samples.To address these issues,this paper proposes a novel spatial scene reconstruction framework.Specifically,the BERT based spatial information knowledge graph extraction method is used to encode the input text,label and classify the encoded text,identify spatial objects and relationships in the text,and accurately extract spatial information.Additionally,a large number of manual experiments were conducted to explore quantitative biases in human spatial cognition,and based on the obtained biases,a greedy resolution method based on cost functions was used to fine tune the layout of conflicting spatial objects and solve the conflicting spatial information in the spatial information knowledge graph.Finally,use graph convolutional neural networks to obtain scene knowledge graph embeddings that consider spatial constraints.In addition,a high-quality training sample set of“text-scene-knowledge graph”was constructed.
