Deep oil and gas resources with multiple faults are the key areas for future energy exploration and exploitation. However, the complexity and attenuation of seismic wave propagation in deep and ultra-deep layers lead to blurred imaging results and low resolution. This paper proposes a method for identifying deep reservoir faults implemented by programming with MATLAB code. The basic process is as follows: Firstly, Gaussian filtering smoothing is performed on the original three-dimensional seismic data; Then extract anisotropic features to define coherent attributes; Then, the edge response is extracted through three-dimensional Laplacian convolution, and the coherent and Laplacian response attribute values are fused. Subsequently, the amplitude of the horizontal gradient is enhanced, while the vertical gradient interference is suppressed through exponential attenuation. Finally, normalization operation is carried out and the contrast is nonlinearly enhanced based on the Sigmoid function to generate the fracture identification body. The application results show that, compared with the traditional three-dimensional seismic fracture identification methods, the method proposed in this paper has better anti-noise interference ability and the ability to distinguish complex fracture features.