In this study, the artificial bee colony (ABC) algorithm was improved to achieve higher accuracy than the latest method in affine (6 DoF) and perspective (8 DoF) template matching. Inheritance of crossover probability and new mutation strategies that improve exploitation are introduced to employed and onlooker bees. In addition, a scout bee that requires no hyperparameters is proposed. Through experiments and quantitative evaluation, we confirmed that the improved ABC achieves more accurate matching and shorter processing time than the latest method based on differential evolution.