With the popularity of crowd sensing, the problem of recruiting the best perceptual information providers at a reasonable cost has become more important, but the greedy nature of mobile users pursuing high returns will make recruitment costs high. For this reason, a method is proposed. The incentive mechanism for group intelligence-aware recruitment is to iterate all possible groups through the attributes and task details of the mobile user. Then, we evaluate the generated random initial group, delete the group that violates the task constraint, and calculate the quality of information(QoI) ratio of the remaining groups. The group will go through the roulette process to select candidates from the current group for the evolution process, the selected group will be crossed, and members will be randomly exchanged between the groups. Finally, the mutation is proceeded, in which the members of the group are replaced randomly, and the group with the best QoI ratio is selected from the solution set to solve the tendency of mobile users overpricing data to increase profits. The proposed incentive mechanism includes selection and payment mechanisms, which avoids the greedy nature of mobile users in the selection process. The effectiveness of the incentive mechanism is proved, by comparing it with the existing team recruitment framework methods, and comparing the experimental data set with the original model.