Abstract: Faced with the problem of large reconstruction errors in three-dimensional pulse signals caused by small displacement of the contact membrane in the existing traditional Chinese medicine fingertip tactile binocular vision detection technology, this study proposes a three-dimensional pulse image detection method based on micro motion amplification technology and explores its application in pulse recognition. Firstly, develop a 3D pulse image detection system based on binocular vision to obtain pulse image signals as experimental data. Then, the phase motion video amplification algorithm is used to amplify the original signal, and the amplified signal is reconstructed in three dimensions to obtain a three-dimensional pulse signal. On this basis, 9 features were extracted from the three-dimensional pulse signal and feature selection was performed using a two sample Kolmogorov Smirnov test. Finally, machine learning algorithms such as decision trees and random forests are used to identify the five types of pulse conditions: sinking, replacement, flood, sliding, and illness. The experimental results show that compared to the absence of micro motion amplification technology, the proposed method significantly improves waveform clarity, amplitude stability, and periodic regularity. Meanwhile, in pulse recognition, the highest average accuracy reaches 96.29 ± 0.258%.