基于自建数据集的比较研究与神经网络实现

With the development of artificial intelligence technology, pattern recognition as one of the important applications has been rapidly developed at home and abroad. Fuzzy pattern recognition is one of the methods based on fuzzy set theory, which can be applied to image classification, face recognition, cell recognition and other fields. In fuzzy pattern recognition, triangle recognition is the most basic pattern recognition, and it is also an important part of practical application problems such as chromosome shape classification and white blood cell classification in the field of cell recognition. However, in triangle fuzzy pattern recognition, there is a lack of datasets available to verify the accuracy of various recognition methods. Therefore, this paper establishes a dataset for triangle fuzzy pattern recognition, including the angle of triangle and the type of identification, to test the accuracy of traditional recognition methods, the recognition method proposed by Cui Xiangjun, the recognition method proposed by Sun Jing and the neural network model. Through experimental results, the application value of this dataset in triangle fuzzy pattern recognition is displayed.