Identifying Dike-Pond System Using an Improved Cascade R-CNN Model and High-Resolution Satellite Images
Open Access
- 3 February 2022
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 14 (3) , 717
Abstract
The dike-pond system (DPS) is the integration of a natural or man-made pond and crop cultivation on dikes, widely distributed in the Pearl River Delta and Jianghan plain in China. It plays a key role in preserving biodiversity, enhancing the nutrient cycle, and increasing crop production. However, DPS is rarely mapped at a large scale with satellite data, due to the limitations in the training dataset and traditional classification methods. This study improved the deep learning algorithm Cascade Region Convolutional Neural Network (Cascade R-CNN) algorithm to detect the DPS in Qianjiang City using high-resolution satellite data. In the proposed mCascade R-CNN, the regular convolution layer in the backbone was modified into the deformable convolutional layer, which was more suitable for learning the features of DPS with variable shapes and orientations. The mCascade R-CNN yielded the most accurate detection of DPS, with an average precision (AP) value that was 2.71% higher than Cascade R-CNN and 11.84% higher than You Look Only Once-v4 (YOLOv4). The area of oilseed rape growing on the dikes accounted for 3.42% of the total oilseed rape planting area. This study demonstrates the potential of the deep leaning methods combined with high-resolution satellite images in detecting integrated agriculture systems.Keywords
Funding Information
- China Agriculture Research System of MOF and MARA (CARS-12)
- Hubei Province Agriculture Research System (HBHZD-ZB-2020-005)
- Fundamental Research Funds for the Central Universities (2662021ZH001)
This publication has 23 references indexed in Scilit:
- Quantification of an overlooked water resource in the tropical rainfed lowlands using RapidEye satellite data: A case of farm ponds and the potential gross value for smallholder production in southern LaosAgricultural Water Management, 2019
- Analyzing Trends of Dike-Ponds between 1978 and 2016 Using Multi-Source Remote Sensing Images in Shunde District of South ChinaSustainability, 2018
- Conversions between natural wetlands and farmland in China: A multiscale geospatial analysisScience of The Total Environment, 2018
- Soft-NMS — Improving Object Detection with One Line of Code2017
- The impacts of integrated homestead pond-dike systems in relation to production, consumption and seasonality in central north BangladeshAquaculture Research, 2017
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Nearest-neighbor diffusion-based pan-sharpening algorithm for spectral imagesOptical Engineering, 2014
- Analysis of Changes in Farm Pond Network Connectivity in the Peri-Urban Landscape of the Taoyuan Area, TaiwanEnvironmental Management, 2012
- A proposal of framework to obtain an integrated biodiversity indicator for agricultural ponds incorporating the simultaneous effects of multiple pressuresEcological Indicators, 2011
- Environmental impact on the development of agricultural technology in China: the case of the dike-pond (‘jitang’) system of integrated agriculture-aquaculture in the Zhujiang Delta of ChinaAgriculture, Ecosystems & Environment, 1996