<p>Traditional feature matching methods rely on accurate descriptors to describe the attributes of feature points, which may lead to the problem of misalignment when the descriptors are incorrect. Considering that correctly matched point pairs should possess the same position, scale, and main orientation after undergoing a transformation model, this paper proposes a feature-constrained registration method for aligning optical and SAR images. In the feature detection stage, a PC_Harris operator is employed, which combines log-Gabor filters with the maximum rectangular phase consistency to detect scale-space features. Compared to traditional multi-scale Harris feature detectors, the PC_Harris feature detector significantly improves the accuracy and stability of feature point matching between optical and SAR images. Furthermore, scale, position, and orientation constraints are introduced in the feature matching process, effectively reducing the problem of misalignment. Experimental results demonstrate its superior matching performance with respect to the state-of-the-art methods.</p>