基于改进EDTER算法的河流边缘检测方法

打开文本图片集

River edge detection based on improved EDTER algorithm

GAO Yudong 1，2 ， HUANG Lingxiao 1，2* ， YAO Xinbo 1，2 ， ZHAO Yuanru³

（1.School of Information Engineering，Ningxia University，Yinchuan 75Oo2l，China; 2. Key Laboratory of Artificial Intelligence and Information Security in the“East Data， West Computing" Initiative ofNingxia，Yinchuan 75OO21，China; 3. School of Civil and Hydraulic Engineering，Ningxia University， Yinchuan 75Oo21，China）

Abstract： To improve the accuracy and robustness of river edge extraction in remote sensing images， a river edge detection method based on an improved EDTER algorithm is proposed. Addressing issues such asedge detail loss and noise interference incomplex natural scenes with traditional methods，a more adaptive edge detection scheme is introduced. The method employs a two-stage edge detection framework with Swin Transformer as the backbone network，incorporating a residual multilayer perceptron to enhance feature representation and designing a lightweight multi-scale decoder to improve detection efficiency.The first stage extracts global image features to generate a global edge map，while the second stage performs local feature extraction and fuses it with globalfeatures to produce the finaledge detection result.Comparative experiments on the BSDS5OO dataset and a self-constructed cross-regional river edge dataset demonstrate that the proposed method outperforms mainstream approaches in ODS，OIS，and AP metrics. Notably， the AP value on the cros-regional river edge dataset reaches O.871，an improvement of approximately 3.2% over the EDTER method， indicating superior edge extraction capability in complex remote sensing scenarios. The proposed method efectively balances detection accuracy and computational effciency， exhibiting strong generalization performance and suitability for water body edge extraction tasks in various remote sensing images，with significant engineering application value.

Key words： river edge；edge detection;remote sensing images； Swin Transformer;waterbody extraction

1引言

河流边缘线作为水体与陆地的分界线，其精准检测对于水资源管理、生态保护及防洪减灾具有重要意义[]。（剩余17749字）