在地震勘探中 ,经过各种处理后最终得到用于解释的地震剖面实际上都可以看作是一些图像。这样 ,就可以从图像处理的观点来看问题 ,图像的特征有边缘和区域 ,那么 ,地震反射同相轴在地震剖面图像上 ,可以看作是一边缘。人工神经网络是近年来发展起来的一种新技术 ,它可以进行二维图像的边缘检测 ,基于这一点 ,把神经网络边缘检测技术应用到地震剖面上来检测地震反射界面 ,压制地震剖面上的随机干扰 ,从而改善地震图像的质量 ,达到提高地震资料信噪比和分辨率的目的。运用该技术对实际剖面进行了处理 ,效果很好 ,证明了方法的可行性和有效性。 In seismic prospecting, the seismic profiles finally obtained for interpretation after various treatments can actually be regarded as some images. In this way, the problem can be seen from the viewpoint of image processing, where the features of the image have edges and regions. Then, the seismic reflection event can be regarded as an edge on the seismic profile image. Artificial neural network is a new technique developed in recent years. It can detect the edge of two-dimensional image. Based on this, the neural network edge detection technology is applied to the seismic profile to detect the seismic reflection interface and suppress the randomness of the seismic profile So as to improve the quality of seismic images and achieve the purpose of improving signal-to-noise ratio and resolution of seismic data. Using this technology to deal with the actual profile, the effect is very good, which proves the feasibility and effectiveness of the method.