Intelligent design of terrain visualization for online maps

doi:10.11947/j.AGCS.2026.20250523

Abstract

Abstract:

Terrain visualization is a core component of map services for representing landforms and surface characteristics. However, existing terrain visualization methods are often highly specialized and poorly adapted to diverse application scenarios, which limits their ability to meet the growing demands of map services. This study proposes a multi-agent 3D terrain visualization method, in which the terrain design process is decomposed into a hillshade agent (HSAgent), a color-relief agent (CRAgent), and a evaluation agent (EAgent). The proposed method supports user intent analysis, cartographic strategy reasoning, iterative self-optimization, and online visualization output in a coordinated workflow. Case tests were carried out in preset scenes and free scenes, covering generative and revised tasks, and subjective visual evaluation and scene matching questionnaire survey were carried out for verification. The results show that this method can realize the rapid design of terrain style, which is superior to the style generated by ArcGIS Pro in visual performance and functional support, and has good generalization effect in different scenarios. It provides an efficient and versatile technical path for online terrain visualization.

Key words: terrain visualization, online maps, multi-agent, intelligent mapping

CLC Number:

P283

Xini HU, Mengjun KANG, Liang GE, Shiliang SU, Min WENG. Intelligent design of terrain visualization for online maps[J]. Acta Geodaetica et Cartographica Sinica, 2026, 55(4): 739-752.

Figures/Tables 18

Fig. 1

Tab. 1

Tab. 2

Tab. 3

Tab. 4

Fig. 2

Tab. 5

Fig. 3

Fig. 4

Tab. 6

Fig. 5

Fig. 6

Fig. 7

Tab. 7

Fig. 8

Fig. 9

Fig. 10

Tab. 8

References 49

[1]	HORN B K. Hill shading and the reflectance map[J]. Proceedings of the IEEE, 1981, 69(1): 14-47.
[2]	KENNELLY P J. Terrain maps displaying hill-shading with curvature[J]. Geomorphology, 2008, 102(3/4): 567-577.
[3]	江文萍, 毋河海, 杜清运, 等. 大型彩色地貌晕渲图的自动生成技术研究[J]. 武汉大学学报(信息科学版), 2004, 29(6): 496-499.
	JIANG Wenping, WU Hehai, DU Qingyun, et al. Automatic generation technology of large-scale colored hill-shading[J]. Geomatics and Information Science of Wuhan University, 2004, 29(6): 496-499.
[4]	李凤霞, 王欣欣, 黄天羽. 一种动态地形实时可视化系统框架[J]. 系统仿真学报, 2009, 21(S1): 151-154.
	LI Fengxia, WANG Xinxin, HUANG Tianyu. A real-time visualization system framework of dynamic terrain[J]. Journal of System Simulation, 2009, 21(S1): 151-154.
[5]	RUZINOOR C M, SHARIFF A R M, PRADHAN B, et al. A review on 3D terrain visualization of GIS data: techniques and software[J]. Geo-spatial Information Science, 2012, 15(2): 105-115.
[6]	JEROME O, BASSEY C, OKEME C, et al. MeshNav3D: software for visualizing and benchmarking uneven terrain planning algorithms[J]. Journal of Open Research Software, 2025, 13: 19.
[7]	郭韵玲, 龚玉叶, 蒋巍, 等. 基于三维地形优化的水源保护区划定研究[J]. 现代农业研究, 2024, 30(12): 112-115.
	GUO Yunling, GONG Yuye, JIANG Wei, et al. Study on the delineation of water source protection area based on 3D terrain optimization[J]. Modern Agriculture Research, 2024, 30(12): 112-115.
[8]	顾绍鹏. 基于Vega Prime/Creator导弹飞行轨迹模拟跟踪技术研究[D]. 哈尔滨: 哈尔滨工程大学, 2016.
	GU Shaopeng. Research on missile flight trajectory simulation and tracking technology based on vega prime/creator[D]. Harbin: Harbin Engineering University, 2016.
[9]	OLYAZADEH R, SUDMEIER-RIEUX K, JABOYEDOFF M, et al. An offline-online Web-GIS Android application for fast data acquisition of landslide hazard and risk[J]. Natural Hazards and Earth System Sciences, 2017, 17(4): 549-561.
[10]	YU Miao, HUANG Yu, XU Qiang, et al. Application of virtual earth in 3D terrain modeling to visual analysis of large-scale geological disasters in mountainous areas[J]. Environmental Earth Sciences, 2016, 75(7): 563.
[11]	DÜBEL S, RÖHLIG M, TOMINSKI C, et al. Visualizing 3D terrain, geo-spatial data, and uncertainty[J]. Informatics, 2017, 4(1): 6.
[12]	杨乃, 孔凡秋, 杨鸿海, 等. 基于空间认知的三维地图光影模型[J]. 测绘学报, 2012, 41(2): 302-308.
	YANG Nai, KONG Fanqiu, YANG Honghai, et al. Model of light and shadow in 3D map based on spatial cognition[J]. Acta Geodaetica et Cartographica Sinica, 2012, 41(2): 302-308.
[13]	ROSE T J, BAKAOUKAS A G. Algorithms and approaches for procedural terrain generation-a brief review of current techniques[C]//Proceedings of 2016 International Conference on Games and Virtual Worlds for Serious Applications. Barcelona: IEEE, 2016: 1-2.
[14]	FERRARIS J, GATZIDIS C. A rule-based approach to 3D terrain generation via texture splatting[C]//Proceedings of 2009 International Conference on Advances in Computer Entertainment Technology. Athens: ACM Press, 2009: 407-408.
[15]	陈望婷, 朱庆. 地形三维可视化中的彩色地貌晕渲图多模式色彩设计[J]. 测绘信息与工程, 2006, 31(6): 47-49.
	CHEN Wangting, ZHU Qing. Design of multi-model colour in colored hill-shading for terrain 3D visualization[J]. Journal of Geomatics, 2006, 31(6): 47-49.
[16]	CHEN W, ZHU Q. Design of multi-model colour in colored hill-shading for terrain 3D visualization[J]. Journal of Environmental Planning and Management, 2007, 50(1): 49.
[17]	林松, 程维明, 乔玉良. 基于Global Mapper的地貌晕渲制图：以西安幅(I-49)为例[J]. 地球信息科学学报, 2009, 11(6): 802-808.
	LIN Song, CHENG Weiming, QIAO Yuliang. Methodology of automatic relief shading map making based on global mapper: taking Xi'an Region (I-49) as an example[J]. Journal of Geo-information Science, 2009, 11(6): 802-808.
[18]	BRATKOVA M, SHIRLEY P, THOMPSON W B. Artistic rendering of mountainous terrain[J]. ACM Transactions on Graphics, 2009, 28(4): 1-17.
[19]	RUSINKIEWICZ S, BURNS M, DECARLO D. Exaggerated shading for depicting shape and detail[J]. ACM Transactions on Graphics, 2006, 25(3): 1199-1205.
[20]	DÜBEL S, SCHUMANN H. Visualization of features in 3D terrain[J]. ISPRS International Journal of Geo-Information, 2017, 6(11): 357.
[21]	KENNELLY P J, STEWART A J. General sky models for illuminating terrains[J]. International Journal of Geographical Information Science, 2014, 28(2): 383-406.
[22]	俞键, 杨世保, 赵宝奇, 等. 基于改进Blinn-Phong光照模型的机载SVS地形着色方法[J]. 电光与控制, 2024, 31(2): 92-97.
	YU Jian, YANG Shibao, ZHAO Baoqi, et al. A terrain shading method of airborne SVS based on improved Blinn-Phong illumination model[J]. Electronics Optics & Control, 2024, 31(2): 92-97.
[23]	GRIGNARD A, FANTINO G, LAUER JW, et al. Agent-based visualization: a simulation tool for the analysis of river morphosedimentary adjustments[C]//Procedings of 2015 International Workshop on Multi-Agent Systems and Agent-Based Simulation. Cham: Springer, 2016: 109-120.
[24]	LI Xiong, PU Wei, RONG Jiang, et al. Terrain visualization information integration in agent-based military industrial logistics simulation[J]. Journal of Industrial Information Integration, 2022, 25: 100260.
[25]	杨必胜, 陈一平, 邹勤. 从大模型看测绘时空信息智能处理的机遇和挑战[J]. 武汉大学学报(信息科学版), 2023, 48(11): 1756-1768.
	YANG Bisheng, CHEN Yiping, ZOU Qin. Opportunities and challenges of spatiotemporal information intelligent processing of surveying and mapping in the era of large models[J]. Geomatics and Information Science of Wuhan University, 2023, 48(11): 1756-1768.
[26]	张帅豪, 潘志刚. 遥感大模型：综述与未来设想[J]. 遥感技术与应用, 2025, 40(1): 1-13.
	ZHANG Shuaihao, PAN Zhigang. Remote sensing large models: review and future prospects[J]. Remote Sensing Technology and Application, 2025, 40(1): 1-13.
[27]	吴华意, 沈张骁, 侯树洋, 等. 大语言模型驱动的GIS分析：方法、应用与展望[J]. 测绘学报, 2025, 54(4): 621-635. DOI: . doi: 10.11947/j.AGCS.2025.20240468
	WU Huayi, SHEN Zhangxiao, HOU Shuyang, et al. Large language model-driven GIS analysis: methods, applications, and prospects[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(4): 621-635. DOI: . doi: 10.11947/j.AGCS.2025.20240468
[28]	李志林, 徐柱, 慎利, 等. 自主式情境化地图表达：大模型时代的智能化地图制图理论探讨[J]. 测绘学报, 2024, 53(11): 2043-2052. DOI: . doi: 10.11947/j.AGCS.2024.20240222
	LI Zhilin, XU Zhu, SHEN Li, et al. Autonomous situatedness map representation: a theoretical discussion on intelligent cartography in the era of large models[J]. Acta Geodaetica et Cartographica Sinica, 2024, 53(11): 2043-2052. DOI: . doi: 10.11947/j.AGCS.2024.20240222
[29]	KANG Yuhao, GAO Song, ROTH R E. Artificial intelligence studies in cartography: a review and synthesis of methods, applications, and ethics[J]. Cartography and Geographic Information Science, 2024, 51(4): 599-630.
[30]	CHENZHEN S, TIAN L, ZHIWEI W U, et al. Generative artificial intelligence and its applications in cartography and GIS: an exploratory review[J]. Journal of Geodesy and Geoinformation Science, 2025, 8(2): 74-89.
[31]	GUO Taicheng, CHEN Xiuying, WANG Yaqi, et al. Large language model based multi-agents: a survey of progress and challenges[EB/OL]. [2025-12-01]. https://arxiv.org/abs/2402.01680.
[32]	CHEN Shuaihang, LIU Yuanxing, HAN Wei, et al. A survey on LLM-based multi-agent system: recent advances and new frontiers in application[EB/OL]. [2025-12-01]. https://arxiv.org/abs/2412.17481.
[33]	ZHANG Yifan, HE Zhengting, LI Jingxuan, et al. MapGPT: an autonomous framework for mapping by integrating large language model and cartographic tools[J]. Cartography and Geographic Information Science, 2024, 51(6): 717-743.
[34]	ZHANG Yifan, WEI Cheng, WU Shangyou, et al. GeoGPT: understanding and processing geospatial tasks through an autonomous GPT[EB/OL]. [2025-12-01]. https://arxiv.org/abs/2307.07930.
[35]	王孟琪, 李伯钊, 王振立, 等. 一种集成知识图谱和大语言模型的智能地图制图框架[J]. 武汉大学学报(信息科学版). [2025-12-01]. https://doi.org/10.13203/j.whugis20240266.
	WANG Mengqi, LI Bozhao, WANG Zhenli, et al. An automatic cartography framework integrating knowledge graph and large language model[J]. Geomatics and Information Science of Wuhan University. [2025-12-01]. https://doi.org/10.13203/j.whugis20240266.
[36]	WANG Chenglong, KANG Yuhao, GONG Zhaoya, et al. CartoAgent: a multimodal large language model-powered multi-agent cartographic framework for map style transfer and evaluation[J]. International Journal of Geographical Information Science, 2025, 39(9): 1904-1937.
[37]	LI Xinyi, WANG Sai, ZENG Siqi, et al. A survey on LLM-based multi-agent systems: workflow, infrastructure, and challenges[J]. Vicinagearth, 2024, 1: 9.
[38]	WOO M, NEIDER J, DAVIS T, et al. OpenGL programming guide: the official guide to learning OpenGL, version 1.2[M]. Boston: Addison-Wesley Longman Publishing Co., Inc., 1999.
[39]	AKENINE-MOLLER T, HAINES E, HOFFMAN N. Real-time rendering[M]. Natick: CRC Press, 2019.
[40]	BREWER C. Designing better maps: a guide for GIS users[M]. 2nd ed. Redlands: ESRI Press, 2016.
[41]	MAPLIBRE C. Layers—MapLibre style specification[EB/OL]. [2025-12-01]. https://maplibre.org/maplibre-style-spec/layers/#paint-hillshade-hillshade-method.
[42]	PATTERSON T, JENNY B. The development and rationale of cross-blended hypsometric tints[J]. Cartographic Perspectives, 2011(69): 31-46.
[43]	WEI J, WANG Xuezhi, SCHUURMANS D, et al. Chain-of-thought prompting elicits reasoning in large language models[C]//Proceedings of 2022 Advances in Neural Information Processing Systems. New Orleans: Neural Information Processing Systems Foundation, Inc., 2022: 24824-24837.
[44]	BUCKINGHAM B, HARROWER M. The role of color saturation in maps for children[J]. Cartographic Perspectives, 2007(58): 28-47.
[45]	FILIPPAKOPOULOU V, MICHAELIDOU E, NAKOS B. A study of children's perception of cartographic landform representation[C]//Proceedings of 1998 Maps for Special Users. Wroclaw: ICA, 1998: 93-104.
[46]	GDAL C. GDALDEM—GDAL documentation[EB/OL]. [2026-02-24]. https://gdal.org/en/stable/programs/gdaldem.html.
[47]	HARROWER M, BREWER C A. ColorBrewer.org: an online tool for selecting colour schemes for maps[J]. The Cartographic Journal, 2003, 40(1): 27-37.
[48]	SAMSEL F, TURTON T L, WOLFRAM P, et al. Intuitive colormaps for environmental visualization[C]//Proceedings of 2017 Workshop on Visualisation in Environmental Sciences. Goslar: Eurographics Association, 2017: 55-59.
[49]	CRAMERI F, SHEPHARD G E, HERON P J. The misuse of colour in science communication[J]. Nature Communications, 2020, 11: 5444.

场景	主要用途	设计特点与原则	光照阴影参数建议/示例
自然生态环境	用于生态格局认知、自然环境展示等	地形表达应以自然连续性和整体协调性为主	方法：igor/combined；Exaggeration：0.3~0.6；Illumination-direction：315°；Highlight-color：冷白色（#e6f2ff）；Shadow-color：（蓝灰色）#2a4858；Accent-color：墨绿色（#c8e6c9）
儿童趣味科普	用于儿童辨识地貌，提升儿童对地形的了解和兴趣	阴影对比度控制在中等水平；高光明亮柔和，维持画面友好度	方法：standard；Exaggeration：0.2~0.4；Illumination-direction：315°；Highlight-color：浅米黄（#fcfaee）；Shadow-color：灰棕色（#757061）；Accent-color：棕褐色（#dc852e）
无人机低空经济	用于支撑低空飞行任务的空间认知与航线规划	根据无人机飞行特点突出可用空间与非可用空间的地形差异	方法：standard；Exaggeration：适中（0.25）；Shadow-color=深灰色（#222222）；Highlight-color=灰白色（#eeeeee）

场景	设计特点与原则	高程配色建议/示例
自然生态环境	参考环境相关的连续色带^[48]，低值区突出草地与自然生态特征，高值区逐步过渡至浅黄或棕色
儿童趣味科普	高程分级适中，等级含义清晰，避免细碎分级，要符合儿童的视觉习惯
无人机低空经济	空域飞行范围设定为[0，3000]m，其中近地活动层（非管制空域）
	如[0，120）m、[120，300）m采用精细分级；[300，3000]（管制空域）
	可采用较粗分级，呈现可用空间与非可用空间的颜色层次

指标范畴	评价指标	指标内涵
Hillshade	视觉平衡性	评估阴影整体效果：①夸张度评估：避免极端值。②颜色对比：阴影色与高光色的对比度是否恰当，如避免阴影比高光亮度高的情况
	坡度清晰度	检查accent-color与阴影/高光的亮度差异是否合适，避免强调色无效（与高光/阴影太接近）
	光源合理性	光源方向、亮部/暗部的色调与亮度关系是否自然可信，如避免冷色亮部、暖色暗部的情况
Color-relief	层级辨识度	层级辨识度基于色带的分段数量与高程覆盖范围进行估计。若覆盖范围较广但是分段数量极少，则分数偏低
	色彩协调度	相邻颜色RGB距离是否适中，避免存在差异过小或过大的极端情况，距离过低（区分不明显）或过高（色彩跳跃）都会降低分数
	色彩平滑度	基于相邻色段在CIELab空间中的色差（ΔE）进行估计，使配色更符合人眼感知^[49]
冲突检测	配色冲突检测	阴影颜色与色带色彩是否出现极近或近互补的色相冲突
冲突检测	不透明度冲突检测	色带不透明度是否过高导致阴影信息被遮挡

模型	参数规模	推理精度	成本与部署难度
DeepSeek-V3-Chat	总计6.71×10¹¹，激活约3.7×10¹⁰	FP16/INT8	成本低，可稳定调用，部署难低
GPT-4	未公开	FP16	调用成本高，本地部署难度大
Llama 3.1	7.0×10¹⁰（全部激活）	FP16/INT8	成本低，受硬件影响大，以英文语料库为主

场景	用户提示词
无人机低空经济	生成一幅无人机低空经济地形样式图
儿童趣味科普	设计一个适合儿童趣味认知的地形样式
城市地形	设计一个清新淡雅的城市地形底图