基于匹配追踪的RGB融合技术及在河道刻画中的应用

摘要

The exact and efficient acquisition of frequency information of seismic data is the key to the time-frequency technique. Compared with common wavelet transform and generalized S transform method, the matching pursuit algorithm can obtain precise position both in time and frequency domain because of its higher time-frequency resolution and local adaptability. The accurate information of frequency domain can help to do seismic attributes and hydrocarbon detection work. RGB plot is a display technique, which divides several frequency data into lower frequency, middle frequency and higher frequency data, with the display fused by red, green, and blue. This technique can reflect information of various bands, which avoids the less information of a single attribute. In practical application, RGB plot by matching pursuit algorithm is used to depict the channel boundary and characterize the heterogeneity of inner channel to a certain degree, which has profound significance in comparison with common attribute.%准确、高效地获取地震资料的频率信息是时频技术的关键。目前最热门的匹配追踪算法相比常用的小波变换和广义 S 变换来说，具有更高的时频分辨率和局部自适应性，能同时在时间域和频率域获得较准确的定位，提取较准确的频率域信息，为地震属性研究、油气检测等研究工作服务。RGB 色彩融合技术是一种显示技术，它可以将多个频率体分为低、中、高三个频带进行融合显示，全面地反映各频带信息，弥补单一属性包含信息较少的缺点。将采用匹配追踪的RGB色彩融合技术应用于实际地震资料，清楚地刻画了河道的边界，并对河道内部的非均质性有一定程度的反映，较常规属性具有更深刻的意义。