Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Research on Low Altitude Object Detection Based on Deep Convolution Neural Network
The rapid and accurate detection of low altitude objects means a great deal to flight safety in low altitude airspace; however, low altitude object detection is very challenging due to the images’ characteristics such as scale variations, arbitrary orientations, extremely large aspect ratio, and so on. In recent years, deep learning methods, which have demonstrated remarkable success for supervised learning tasks, are widely applied to the field of computer vision and good results have been achieved. Therefore, the deep learning method is applied to low altitude object detection in this paper. We proposed a deep convolution neural network model, which utilizes deep supervision implicitly through the dense layer-wise connections and combines multi-level and multi-scale feature. The model has achieved state-of-the-art performance on two large-scale publicly available datasets for object detection in aerial images.
Research on Low Altitude Object Detection Based on Deep Convolution Neural Network
The rapid and accurate detection of low altitude objects means a great deal to flight safety in low altitude airspace; however, low altitude object detection is very challenging due to the images’ characteristics such as scale variations, arbitrary orientations, extremely large aspect ratio, and so on. In recent years, deep learning methods, which have demonstrated remarkable success for supervised learning tasks, are widely applied to the field of computer vision and good results have been achieved. Therefore, the deep learning method is applied to low altitude object detection in this paper. We proposed a deep convolution neural network model, which utilizes deep supervision implicitly through the dense layer-wise connections and combines multi-level and multi-scale feature. The model has achieved state-of-the-art performance on two large-scale publicly available datasets for object detection in aerial images.
Research on Low Altitude Object Detection Based on Deep Convolution Neural Network
Stud. in Distributed Intelligence
Yuan, Xiaohui (Herausgeber:in) / Elhoseny, Mohamed (Herausgeber:in) / Qi, Yongjun (Autor:in) / Gu, Junhua (Autor:in) / Tian, Zepei (Autor:in) / Feng, Dengchao (Autor:in) / Su, Yingru (Autor:in)
26.06.2020
10 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
A Convolution–Non-Convolution Parallel Deep Network for Electricity Theft Detection
| DOAJ | 2023
Intelligent crack detection based on attention mechanism in convolution neural network
| SAGE Publications | 2021