【Objective】Viruses play a crucial role in regulating soil microbial community and host dynamics. However, the efficient extraction and enumeration of soil viruses remain challenging due to the complexity of soil environments. This study aimed to enhance the efficiency of soil virus detection by optimizing both the viral extraction protocol and epifluorescence microscopy observation conditions. 【Method】Soil samples were collected from the long-term brown soil fertilization experimental station at Shenyang Agricultural University. Three key factors affecting soil virus extraction efficiency and enumeration results were evaluated: (1) antifade agents (the commercial anti-quenching agent Fluoromount-G? and a laboratory-prepared antifade agent); (2) disruption methods (blender, vortex mixer, ultrasonic cell disruptor, and water bath shaker); and (3) soil-to-liquid ratios (30:100 and 50:100; mass: volume). Viral particles were stained using SYBR Green I and enumerated under an epifluorescence microscope. The optimized protocol was subsequently applied to brown soil, black soil, and meadow soil to validate its general applicability. 【Result】Without an antifade agent (Fluoromount-GTM or laboratory-prepared antifade agent), viral particles were difficult to observe under 100× magnification. The application of Fluoromount-GTM significantly increased virus counts to 3.75 × 108 virus-like particles (VLPs) g-1, outperforming the laboratory-prepared antifade agent. Disruption method significantly affected virus extraction efficiency (P < 0.05). The blender (650 W, 50 Hz) treatment for 3 minutes achieved the highest extraction efficiency, significantly surpassing the vortex mixer and water bath shaker, whereas no viral particles were observed with the ultrasonic cell disruptor treatment. Increasing the soil-to-liquid ratio also significantly enhanced virus extraction efficiency (P < 0.05), with the 50:100 ratio resulting in a 1227.10% increase in virus counts compared to the 30:100 ratio. Validation results demonstrated that the optimized method consistently yielded stable and reliable viral counts across all three soil types, with significant differences in viral abundance observed among them, confirming its broad applicability. 【Conclusion】In summary, this study established an efficient, stable, and widely applicable epifluorescence microscopy method for soil virus enumeration through the systematic optimization of the disruption method, soil-to-liquid ratio, and antifade agent. The optimized method significantly enhances the extraction efficiency and counting accuracy of soil viruses, thereby providing robust technical support for soil viral ecology research.