Deep Belief Network And Its Application In Handwritten Font Recognition

Deep learning is currently an extremely active research area in machine learning and pattern recognition society. Deep learning and Big Data are two hottest trends in the rapidly growing digital world. Today big data for the development of various industry has brought the huge opportunity a nd potential, on the other hand also brought unprecedented challenges. While Big Data has been defined in different ways, herein it is referred to the exponential growth and wide availability of digital data that are difficult or even impossible to be managed and analyzed using conventional software tools and technologies. It has gained huge successes in a broad area of applications such as speech recognition、computer vision、and natural language processing. In the field of computer vision optical character recognition(OCR) concept was put forward in the early 1920 s, is a representative in the field of pattern recognition research important topic. In recent years, with the development of mobile Internet, optical character recognition technology application scenario is also from the traditional office areas such as postal code, scan books, handwritten gradually penetrate into the daily life, produce many cellphone camera optical character recognition as the entrance to the text information input and query applications.In this paper, we based on neural network, the interpretation of the concept of deep learning system. Boltzmann Machines is the basis of deep learning theory, Boltzmann Machines in dimension reduction, classification, collaborative filtering is used in the modeling, study and subject characteristics, such as the paper to recognize handwritten font depth belief network building, all need to limit the Boltzmann Machines for feature extraction.So the theory and practical application of Boltzmann Machines has a long value, this paper mainly described the current situation of the development of Boltzmann Machines, mainly includes the Restricted Boltzmann Machines network structure and the classification of network topology, the basis of limit theory, the basis of Boltzmann Machines is described in Restricted Boltzmann Machines learning process, Restricted Boltzmann Machines main purpose.Then used Restricted Boltzmann Machines to construct built Deep Belief Networks., respectively, using the BP neural network and deep belief network used for handwriting recognition, according to the recognition result is analyzed the advantages and disadvantages of two kinds of algorithm. In the end, concludes with a discussion and some plans for future work.