Distinct lines of illumination produced by first and second lasers are swept across a field of view using a scanning mirror, illuminating the field of view with a known time delay. The second laser may be physically offset by a fractional pixel distance relative to the first laser. Reflections from the field of view are detected by corresponding detectors, and a processing unit can interleave pixels generated from the distinct illuminations to create a higher-resolution combined point cloud. A pixel-by-pixel radial velocity can be calculated based on a change in radial distance between corresponding pixels over the known time delay. Edge artifacts, such as anomalies caused by laterally moving objects, are detected and filtered by comparing the pixel-level radial velocities to a predefined threshold limit. The remaining valid pixels can then be grouped into recognized objects to calculate a velocity of the object.