Moving target imaging(MTI) plays an important role in practical applications. How to capture dynamic images of the targets with high qualities has become a hot point of research in the field of MTI. In order to improve the reconstruction quality, a new MTI model based on compressed sensing(CS) is proposed here, by using a sampling protocol of the row-scanning together with a motion measurement matrix constructed by us. It is proved by the simulation and the experimental results that a relatively high quality can be achieved through this approach. Furthermore, an evaluation criterion of reconstructed image is introduced to analyze the relationship between the imaging quality and the moving speed of the target. By contrast, the performance of our algorithm is much better than that of traditional CS algorithm under the same moving speed condition. As a result, it is suggested that our imaging method may have a great application prospect in the earth observation of unmanned aerial vehicles, video monitoring in the product line and other fields.