In order to solve the pattern recognition problem that the index value and weight are both interval grey numbers, a multi-index grey pattern recognition model based on generalized greyness is established in this paper. Firstly, the basic concepts of grey pattern, universal grey pattern and grey pattern recognition are given, and the principle of maximum possibility degree and the possibility function of single index grey pattern are given. Secondly, two methods of multi-index grey pattern recognition are given, that is, the comprehensive possibility degree model and proximity degree model, and the simplified forms of proximity degree model are given. Finally, the grey pattern recognition model is applied to the high-quality development evaluation of 15 national agricultural science and technology parks in Shandong province of China to verify the validity of the model. The results show that the proposed grey pattern recognition model has the advantages of strict mathematical foundation, clear physical meaning, simple calculation and easy programming. The application examples show that the proposed grey pattern recognition model is feasible and effective. The research results not only enrich the theory of grey mathematics and grey system, but also provide a new way for grey pattern recognition in cases where the weights and index values are all interval grey numbers or the coexistence of interval grey number and real number.