Using a microwave or millimetre wave radar system, this study aims to achieve highly accurate Doppler velocity estimation and radar imaging that would be suitable for various remote-sensing sensors, such as self-driving, surveillance, or security applications. In particular, micro-Doppler signatures are one of the most promising approaches for human recognition; however, the traditional limitations of both velocity and temporal resolution need to be addressed. The weighted kernel density (WKD) algorithm using super-resolution Doppler velocity estimation has been one solution. Using WKD, this study incorporates the range points migration (RPM) method of radar imaging to enhance accuracy in both Doppler velocity and radar imaging using an iterative data selection scheme. Furthermore, to obtain more informative Doppler-associated RPMbased images using less data by exploiting a unique feature of WKD and RPM methods, this study introduces the image integration approach along the slow-time profile. In this framework, bi-directional data processing between Doppler velocity and imaging analysis is achieved. At each pulse hit sequence, both numerical and experimental tests demonstrate that the proposed method yields a more accurate Doppler-associated radar image compared with the methods in previous studies.