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.
Keywords clustering, as the basic method of domain knowledge analysis, has some problems such as difficult to understand the clustered tree diagram, scarce of further analysis methods, etc. The paper proposed a new approach to analyze domain knowledge based on keywords clustering. The proposed weighted knowledge model (WKN) is composed of two types of nodes (nodes of high-frequency keywords and nodes of clusters which come from keywords clustering and named as keywords nodes). Based on WKN, some new methods are suggested to analyze domain knowledge, such as main sub-fields analysis and representation, important sub-fields and hot spots of domain knowledge identification, research fronts analysis, etc., and all the analysis results can be illustrated as a sub-network of WKN. In the end, a case study was conducted to verify the feasibility and validity of the methods. Compared with the existing methods, the proposed methods seem more clearly, deeply and conveniently, and present new tools for researchers to study and utilize domain knowledge.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.