Abstract With the rapid development of digital graphic recognition technology, in recent years, more attention is given to the subject of animal and plant species identification which has become a popular researching field. The article mainly discusses about the method to apply graphic analyzing technology for bird identification research in the case of acquatic bird protection in the marsh of Minjiang River estuary. It also put forward a set of classification case with SVM Decision Tree for protesting information through deep analysis and study on bird color graph  in terms of shape, color and texture. The case takes on many advantages such as, excellent classification capacity, firm Over-generalization and accurate classification, which help to solve the disturbing classification problem lasting for many years. Moreover, the appearance of the case contributes to the effective recognition and classification of 15 birds species in Minjiang River estuary wetland.