基于VIIRS数据的油气平台提取技术研究

发布时间：2018-03-23 11:20

本文选题：卷积运算临界值法　切入点：油气平台　出处：《地球信息科学学报》2017年03期

【摘要】：油气平台作为海上油气资源勘探开发的主要设备之一,其数量与空间分布反映了一个区域油气资源的开发状况。普通光学影像易受天气状况影响,而雷达数据成本较高,这给油气平台的检测带来了一定的困难。由于油气平台作业过程中,需要灯光照明,同时伴气的燃烧也产生很强的灯光,因此可以通过检测灯光来实现油气平台的提取。针对VIIRS数据具有强夜间光探测能力,本文提出了一种卷积运算临界值法对海上油气平台进行提取。首先对2期不同时相的VIIRS数据进行卷积运算,对亮像元进行了增强处理,对背景像元进行弱化处理,从而明确区分疑似目标与背景,然后以0值为分界点,对疑似目标进行提取,最后利用油气平台相对静止的特性,通过邻域分析,实现油气平台的提取。结果表明,本文提出的卷积运算临界值法可以有效地提取油气平台,提取准确率约为85.4%,同时可以有效地减少由于经验阈值对提取结果造成的误差。
[Abstract]:As one of the main equipment for offshore oil and gas exploration and development, the quantity and spatial distribution of oil and gas platform reflect the development of oil and gas resources in a region. This brings some difficulties to the detection of oil and gas platforms. During the operation of oil and gas platforms, lighting is needed, and the combustion of gas companion also produces very strong lights. Therefore, the oil and gas platform can be extracted by detecting the light. For the VIIRS data, it has a strong night-time light detection ability. In this paper, a critical value method of convolution operation is proposed to extract offshore oil and gas platform. Firstly, convolution operation is carried out on the VIIRS data of two periods of different phases, the bright pixel is enhanced, and the background pixel is weakened. In this way, the suspected target is clearly distinguished from the background, and then the suspected target is extracted by using zero as the boundary point. Finally, the extraction of the oil and gas platform is realized by neighborhood analysis using the characteristics of the relatively static oil and gas platform. The convolution critical value method proposed in this paper can effectively extract the oil and gas platform, and the extraction accuracy is about 85.4. At the same time, it can effectively reduce the error caused by the empirical threshold.
【作者单位】：兰州交通大学测绘与地理信息学院;中国科学院地理科学与资源研究所资源与环境信息系统国家重点实验室;中国南海研究协同创新中心;
【基金】：高分辨率对地观测系统重大专项(00-Y30B15-9001-14/16-5)“重要战略资源(石油)安全动态监测评估子系统及应用示范(一期)”
【分类号】：TE53

【参考文献】