Global Navigation Satellite System (GNSS) spoofing is nowadays an emerging topic, especially in safety-critical applications. Power monitoring and signal quality monitoring (SQM) are effective ways to identify spoofing by monitoring the abnormal power and the distorted auto-correlation function (ACF). However, the matched power spoofing is difficult to detect, and the SQM technique is only effective in the spoofing drag-off stage. In response to a need for a single metric that combines power monitoring and ACF distortion monitoring, a graphical way of continuous GNSS spoofing detection is proposed in this paper. This paper defines a new metric called ACF similarity to characterize the power of the spoofing signal and the spoofer dragging process. Based on the image features extracted from the time-domain transient response of multiple correlators outputs, the proposed metric can track the ACF and power abnormalities in both spoofer drag-off and steady-state periods. Simulation results prove that the ACF similarity statistic follows Gaussian distribution. As such, a spoofing detection algorithm with an optimal threshold is developed based on the NeymanPearson theorem. The performance of the proposed detector has been verified by utilizing the Texas Spoofing Test Battery (TEXBAT) dataset. Results show that with 1% false alarm rate, the detection probability of the proposed detector achieves 100% in Scenario 5 and 87% in Scenario 3 in TEXBAT dataset.