论文部分内容阅读
针对目前我国社会治安视频(卡口)规划存在的选点凭经验、规划缺指标、防控区域无法量化、成果缺乏科学性等问题。本文提出了一种基于时空大数据挖掘分析理论支撑的视频(卡口)空间规划方法,通过对已建视频(卡口)与用地、建筑、水系、交通、人口、案件、车流、商圈等时空分析,挖掘出与视频(卡口)选址相关的量化参数和指标,基于空间视觉金字塔的层次分析模式建立顾及人、车、案、商圈四要素的视觉金字塔分级选址模型,以最大覆盖模型理论对规划结果进行视域覆盖评价。选择3个城市进行模型实验与算法验证,对实验成果的布点精度和视域覆盖进行了评估。结果表明,模型计算出的空间布点精度中误差m=±13.05米,视频(卡口)视域完全覆盖了治安防控区域,模型科学可靠,可作为治安视频与卡口量化规划参考。
In view of the current selection of social security video (bayonet) planning in our country, there are some problems such as the lack of indicators, the inability to quantify the prevention and control area, the lack of scientific results and so on. This paper presents a method of video (bayonet) spatial planning based on the theory of big data mining and analysis of spatio-temporal space. Through the analysis of video (bayonet) and land use, construction, water system, traffic, population, case, traffic flow, The spatial and temporal analysis is carried out to find out the quantitative parameters and indexes related to video (bayonet) site selection. Based on the hierarchy analysis model of spatial visual pyramid, a visual pyramid hierarchical location model that takes into account the four elements of people, vehicles, cases and business districts is established. Coverage model theory evaluates the coverage of the planning results. Three cities were selected for model experiment and algorithm verification, and the precision and coverage of experimental results were evaluated. The results show that the accuracy of spatial distribution accuracy calculated by the model is m = ± 13.05 meters. The video (bayonet) sight completely covers the area of public order prevention and control. The model is scientific and reliable and can be used as a reference for the quantitative planning of video and bayonet of public security.