Target Recognition Network Based on Improved Convolutional Regression Network

Abstract For refine the regression precision and speed of convolutional neural recognition network, propose improved detection identification algorithm on basis deep learning algorithms, mainly aiming at improving accuracy prediction target recognition. Firstly, acquired images are preprocessed in batches to obtain image datasets with different sizes objects, that is, enhancement small sample images. Secondly, one-stage network SSD is improved. ResNet50 used extract features, while FPN fused main ameliorate which purpose improve accuracy. The test results remote sensing show actual was elevated about 0.9% compared conventional algorithm. measurement also increased by 0.6 frame/s.