Abstract. A nondestructive surface-enhanced Raman spectroscopy (SERS) method based on silver colloids (AgNPs) was developed for rapid detection of carbendazim in apples. SERS which is known for its ultra-sensitivity and rapidity in sample analysis is a powerful analytical technique. This research was conducted based on a self-developed Raman system. The SERS spectra were obtained using 785 nm exciting radiation with 450 mW laser power and 3 s exposure time. Raman spectra of organic substances have severe fluorescence background, so it is crucial to remove fluorescence background from Raman signal first for subsequent signal analysis. In this study, 8 th order polynomial fitting method, min-max signal adaptive zooming method and adaptive iteratively reweighted penalized least squares method were employed to remove the fluorescence background. The intensity of the Raman signal depends on the carbendazim concentration of apples. The peak intensity of carbendazim at 631.8 cm -1 , 1231.3 cm -1 , 1521.7 cm -1 can be used for monitoring the pesticide levels. A good linearity (R=0.973) was found in the concentration range of 0.014 mg/kg to 1.34 mg/kg in apples. The lowest detectable level for carbendazim in apples was 0.014 mg/kg, which was sensitive enough for identifying apple contaminated with carbendazim above the maximum residue level. The present study demonstrated a novel approach to detect carbendazim residues in apples by using AgNPs and the results indicated a great potential of the proposed method for the evaluation and identification of pesticide residues in agricultural products.