The impulse response method using a laser Doppler vibrometer (LDV) was performed to nondestructively measure pear quality. To get a wide range of texture and different freshness in pears, the experiment was conducted every other day during 7 days storage. Each pear was excited by a half-sine impulse signal, and an LDV was used to measure the response signal from the top of the pear. A fast Fourier transform algorithm was used to transform time domain signals to frequency domain signals. A total of 15 and 8 features were extracted from the time and frequency domain signals, respectively. Pear texture was measured by the puncture test. Maximum force (ME), flesh firmness (FF) and stiffness (Stif) were extracted from the force-deformation curve as texture indices. Different modeling methods, including the stepwise multiple linear regression (SMLR), back propagation neural network (BPNN), and principal component analysis-back propagation neural network (PCA-BPNN) methods, were used for quantitative analysis of pear texture. Best prediction results were obtained by the PCA-BPNN method, especially for predicting FF with correlation coefficient (r(p)) of 0.840 and root mean square error of prediction (RMSEP) of 0.959 N. The Fisher's discriminant analysis (FDA), BPNN, and PCA-BPNN methods were applied to qualitative analysis of pear freshness. Pears were categorized into 4 groups with different freshness according to the 4 test days. The best results were also obtained by the PCA-BPNN method, resulting in accuracy of 89.0% and 83.3% for calibration and validation, respectively. Experimental results showed that the impulse response method using an LDV is capable for evaluating pear texture and freshness. The proposed approach provides a way for rapid detection of pear quality to meet the requirement of on-line detection. (C) 2015 Elsevier Ltd. All rights reserved.