GloVe-FRCNN: Comprehensive network algorithm for Vespa mandarinia image-text extraction and classification
Shien Sun, Zheng Gao, Chenyu Huang, Haitao Yu · 2021
In this paper, a synthesis algorithm of text classification and image recognition is proposed. This algorithm directly extends the original image recognition method. And its novelty is based on the fusion algorithm of image recognition and text classification as a new mathematical derivation approach, which uses stepwise linear fitting to fuse multiple models. We use the new fitting method, which makes use of the idea of fitting the model with different weights, to expand the training data. The principle of the method is introduced in detail, and the experimental results are given.