Rate control plays a crucial role in controlling compression bitrates and encoding qualities for networked video applications. The intraframe quantization parameter is very important for enhancing video quality and retaining a limited bit budget. In this paper, we propose a new total variation (TV) measurement based distortion-quantization model, which is found to be a better measurement of perceived image quality than mean squared error. According to the proposed model, at the frame level, we give a method to decide the intraframe quantization step size by the mean distortion of previous coded frames in the last group of picture (GOP) to improve the coding quality and obtain the stable visual quality. In order to further improve Rate Control performance, based on TV measurement, an initial intraframe quantization step size model and a scene change detection method are also proposed. Experimental results show that, when compared with JVT-W042 rate control algorithm, adopted by H.264/AVC standard, the proposed algorithm achieved more accurate rate control in reducing frame skipping, decreasing quality fluctuation and improving the overall coding quality.