A New Type Method of Adhesive Handwritten Digit Recognition Based on Improved Faster RCNN

Aiming at the low recognition accuracy of the traditional machine learning algorithm which is susceptible to digital writing quality, inter-digital adhesion, random noise background and other factors in the process of adhesion handwritten digit recognition, an new method based on improved fast regional convolutional neural network(Faster RCNN) of adhesion handwritten digit recognition is proposed. Firstly, the NIST19 dataset is used as the basic dataset, and a mixed dataset is created by setting different hand-to-hand ratios with different degrees of overlap, and then randomly add salt and pepper noise and Gaussian noise in the experimental images. Secondly, aiming at the problem of a large number of overlapping objects in the handwritten digital images, a model based on improved Faster RCNN network is built and trained with the above data sets. Finally, the average accuracy of the model is evaluated. The experimental results show that the average detection accuracy of the proposed model is good. Compared with the original Faster RCNN and YOLO models, the improved model not only reduces the scale of parameters, but also ensures high recognition accuracy, and realizes the accurate and efficient recognition of handwritten adhesive digits.