Stereo vision has emerged as a reliable and cost-effective technique for depth perception in intelligent vehicles. In this project, we design and implement a vehicle-mounted stereo distance measurement system that leverages binocular cameras to obtain real-time three-dimensional information of the surrounding environment. The proposed framework consists of several key stages: image acquisition from dual cameras with a fixed baseline, image rectification to align epipolar lines, feature extraction and correspondence matching, disparity computation, and depth estimation using geometric constraints. To enhance measurement accuracy and robustness under complex road conditions, optimized feature matching strategies and filtering algorithms are integrated into the pipeline. The system outputs depth maps and distance measurements that are visualized in a color-coded format and further applied to downstream tasks such as obstacle detection, autonomous navigation, and 3D environment reconstruction. Experimental results demonstrate that the system achieves accurate and stable distance estimation with real-time performance, providing a promising solution for advanced driver assistance systems (ADAS) and autonomous driving platforms.