Vortex flowmeters are easily affected by vibration interference of pipelines and various equipment when used in industrial field. Especially in the measurement of small flow, the noise signal will be superimposed on the output signal of the sensor, the vortex signal is easily submerged, and the measurement of small flow is limited. Aiming at the problem that the small flow vortex signal is very weak and difficult to detect under actual working conditions, an adaptive stochastic resonance (ASR) detection method based on ensemble empirical mode decomposition (EEMD) is proposed. Firstly, the vortex signal is denoised and preprocessed by EEMD, and the appropriate component is selected to reconstruct the signal by using the correlation coefficient as the screening criterion. Then, the system parameters are optimized in parallel by particle swarm optimization to achieve the best stochastic resonance result. The numerical simulation and experimental research results show that this method can improve the output power spectrum amplitude of the vortex signal with small flow, effectively obtain the vortex frequency, realize the measurement of small flow, and enhance the detection ability of weak signal.