Variation Analysis-Based Public Turing Test to Tell Computers and Humans Apart

The present invention shows a gesture verification method, network gesture verification system and verification method. The gesture verification method includes the following steps: S1. A system gives an image containing a randomly generated reference trajectory and presets a minimal match degree; S2. Displaying the image to the user; S3. The user draws an resemblant trajectory to match the reference trajectory; S4. Scaling the resemblant trajectory to adapt to the size of the reference trajectory. S5. Matching the scaled resemblant trajectory with the reference trajectory, if the match degree is lower than the minimal match degree, the verification fails; if the match degree is equal to or higher than the minimal match degree, the verification succeeds. The present invention (VAPTCHA) is safer and more user-friendly than the existing CAPTCHAs, which can be applied to PCs, mobile phones and other embedded devices.