Nowadays, SAR imaging systems have become more interesting for scientists. High spatial resolution of recent SAR images has become a key factor for their accurate applications such as interferometry, radargrammetry and change detection. In these applications, SAR image matching procedure is a critical step for conjugate point extraction which is much more difficult than the optical images. The main reasons are both high radiometric speckle noise and high geometric distortions defining by layover, shadow, and foreshortening. Therefore, utilizing efficient methods to achieve a more favorable result in the SAR image matching, could help us to get better output in various applications. Among the SAR images, study about TerraSAR-X is so important, because TerraSAR-X sensor acquires high spatial resolution images. In this paper, we used various textural image features including contrast, correlation, dissimilarity, entropy, homogeneity, optimal gradient filter, mean, second moment, variance, high pass filter and edge images. The main aim of our research was to determine how much can these features improve the SAR image matching process? In our experiments, we used a small urban part of pair of TerraSAR-X single-look slant-range complex (SSC) images that were acquired over the city of Jam, southern Iran, in spotlight mode. The result shows that for some image areas with low texture, the conventional image matching algorithm (here Lucas-Kanade’s optical flow) is not able to detect corresponding points, while using the above textural features in image matching process causes to extract appropriate numbers of matched points.