Figure 1. Example of video frame extrapolation. Top is the extrapolated result, middle is the zoomed local details and bottom is the occlusion map computed with ground truth.
Dealing with the insufficient detection accuracy and speed of aircraft targets in remote sensing images under complex background, this paper proposes a new detection method, YOLOv5-Aircraft, based on the YOLOv5 network. The YOLOv5-Aircraft model is improved in 3 ways: (1) At the beginning and end of original batch normalization module, centering and scaling calibration are added to enhance the effective features and form a more stable feature distribution, which strengthens the feature extraction ability of network model. (2) The cross-entropy loss function in the confidence of the original loss function is improved to the loss function based on smoothed Kullback-Leibler divergence. (3) For reducing information loss, the CSandGlass module is designed on the backbone feature extraction network of YOLOv5 to replace the residual module. Meanwhile, low-resolution feature layers are eliminated to reduce semantic loss. Experiment results demonstrate that the YOLOv5-Aircraft model can enhance the accuracy and speed of aircraft target detection in remote sensing images while achieving easier convergence.
Text image recognition has opened up a new path for intelligent text sorting and research. The number of epitaph inscriptions is very large and of great value. Due to the limitation of objective conditions, there are many deficiencies in the study of its text arrangement and it is urgent to adopt new methods and new technologies to overcome them. This paper attaches great importance to the design of an automatic recognition algorithm for tablets based on the convolutional neural network. It also describes the application prospects of using the algorithm for the reading and distinguishing of inscriptions, the sorting of inscriptions, and the building of intelligent databases.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.