With the improvement of people’s living standards, people pay more and more attention to the health care function of Chinese herbal medicine. Dendrobium officinale Kimura et Migo (Dendrobium officinale) is a rare Chinese herbal medicine, known as “life-saving fairy grass”. In this study, we tried to evaluate the quality of Dendrobium officinale by sugar content, pH value and other related physical and chemical characteristics. We selected three different habitats of Dendrobium officinale from Huoshan in Anhui, Yandang Mountain in Zhejiang and Yunnan Province as research objects, extracted the spectral data and physical and chemical parameters of different Dendrobium officinale, and then carried out the inversion of quality indicators. Finally, the correlation model between quality and spectrum was established. In the experiment, the leaves, roots and flowers of Dendrobium were removed first, and then the stems to be studied. The spectral data of three Dendrobium officinale with different quality grades were obtained by the ASD specter. The same samples was ground, put into a centrifugal tube, sealed with methanol solution, and packaged with tin foil paper to make corresponding solutions. The chlorophyll content, sugar content and pH value were measured by spectrophotometer, sugar meter and pH meter. The upper layer, middle layer and lower part of the centrifuge tube were selected for each sample. Each sample was measured 3 times, and the average value was taken as the control sample. The original spectral data were denoised and dimensionally reduced by wavelet transform. The correlation between the energy coefficients (including band and scale) and the physicochemical parameters of the Dendrobium officinale control group was analyzed. The higher energy coefficient in the determination coefficient was selected as the wavelet feature, and the wavelet feature was fitted by the least square method. Using all experimental samples as test set and 70% as verification set, the determination coefficient (R²) of chlorophyll content inversion model was 0.819, 0.820 and 0.865, the root mean square error (RMSE) were 0.035, 0.013 and 0.017, respectively; the determination coefficient (R²) of sugar content inversion model was 0.756, 0.764 and 0.823, respectively. The results showed that the root means square error (RMSE) was 0.025, 0.030 and 0.036 8; the determination coefficient (R²) of the inversion model for pH value was 0.819, 0.820 and 0.865, and the root mean square error (RMSE) was 0.034 5, 0.013 and 0.017, respectively. It can be found that the quality inversion model and determination coefficient (R²) of three kinds of Dendrobium officinale are all greater than 0.80, and the root means square error (RMSE) is less than 0.10. This study proved that the spectral characteristics of chlorophyll, sugar content and pH value in Dendrobium officinale were feasible for quality evaluation.