This paper explores the in vivo action of an anti-cancer polypeptide (denoted by: GPG) with low toxicity that targets tumor cells. This aims to overcome the limitations of current anti-cancer therapy, and explores tumor evaluation and mass coverage status with the targeting peptide, by the use of spectrofluorimetry. The study focuses on monitoring the main actions of the anti-cancer peptide in vivo, with only one set of mouse models based on dual reporter dyes (H22 hepatocarcinoma cell transfected with EGFP, denoted by H22-EGFP, and GPG labeled with fluorescent dye Cy7, denoted by Cy7-GPG, were used thereof). A transplanted tumor mouse model was constructed with H22-EGFP and Cy7-GPG was injected in the tail vein. The imager(Ex = 750 nm) revealed that the number of fluorescent photons in the tumor increased continuously from (3.90±0.260)×106 photons·(s·cm2)-1 at the 4th hour (the same unit below) to (1.28±0.330)×108 at the 24th hour. Orange fluorescence accumulated on the tumor completely, with no fluorescence accumulation on the tumor in the Cy7 control group. There was no significant change of fluorescent photon number on the tumor as well. At this time, images of the coating Cy7-GPG and actual tumor mass in the same experiment mouse were recorded, at 750 and 488 nm excitation wavelength respectively, indicating that they were of the same size and shape. Subsequently, GPG was injected into the mice in each of the experimental groups every 2 days. The imager(Ex = 488 nm) revealed the tumor had become smaller, while the number of fluorescent photons gradually decreased from (4.15±0.291)×106 on the 2nd day to (4.75±0.283)×104 on the 56th day. However, the condition of the blank control group was just the opposite. The anti-cancer activity of GPG was comparable to cyclophosphamide. However, the latter had more serious toxic side effects on mice. Finally, mice in the GPG efficacy experiment group were injected with Cy7-GPG and were then sacrificed 48 hours later, with the internal organs and tumor masses extracted for recording the number of fluorescent photons using the imager(Ex=750 nm). The experiment showed that GPG had precise targeting, wider coverage, and strong efficacy, with no toxic or side effects on other organs. The fluorescence imaging model constructed using dual reporter dyes used in this paper was utilized to monitor the coverage of targets; it overcame the limitations of the traditional evaluation method for targeting the function of peptide and improved the understanding of the mechanism of drug action. Only one set of mouse models was used in this experiment. Hence the main performance of the targeting peptide was monitored in a simple and cost-effective manner, indicating that GPG has excellent performance and useful application in vivo.