For the white noise always embeds in signals of gas sensor array, there must be bad influence on the accuracy and reliability of test results while micro or trace organophosphorus pesticide residues in vegetables detection. To solve the problem, acephate and phoxim were selected as test object. A denoising method of gas sensor array signals based on wavelet packet decomposition and reconstruction was practiced. With the help of principal component analysis (PCA) and Fisher discriminant analysis (FDA), the identification results of different concentrations corresponding to the two kinds of pesticide residues were explored and compared, respectively. The results showed that the different concentrations samples corresponding to acephate and phoxim were all well discriminated after the gas sensor array signals treated by the wavelet packet denoising. Therefore, the wavelet packet denoising method could improve identification effect of pesticide residues in vegetables using gas sensor array.