Figure 4 - available via license: CC BY
Content may be subject to copyright.
Source publication
Remote sensing has been widely used in vegetation cover research but is rarely used for intercropping area monitoring. To investigate the efficiency of Chinese Gaofen satellite imagery, in this study the GF-1 and GF-2 of Moyu County south of the Tarim Basin were studied. Based on Chinese GF-1 and GF-2 satellite imagery features, this study has deve...
Context in source publication
Citations
... Intercropping is associated with weed reduction, the control of pests and diseases (Bybee-Finley and Ryan, 2018;Zaefarian and Rezvani, 2016), the preservation of soil nutrients and reduction of rainwater runoff (Sun et al., 2019;Yang et al., 2020). It can also promote biodiversity (Liu and Chen, 2019), sustain food production, and diversify income (Zuo et al., 2013). According to Himmelstein et al. (2017), on average intercropping can increase crop yields by 23 %. ...
... The majority of these studies have explored different classification methods, with machine learning methods being the most preferred approach. Notably, mainly Random Forest (RF) and Support Vector Machine (SVM) have demonstrated superior perfomance in achieving high classification accuracies (Kuchler et al., 2020;Liu et al., 2013;Liu and Chen, 2019). Despite their effectiveness, the results of machine learning methods, including RF and SVM, can be inconsistent (Feyisa et al., 2020). ...
... We chose a Random Forest (RF) classifier as it is known to be superior and reliable for achieving high classification accuracies when compared to other classifiers such as the binary hierarchical classifier, artificial neural networks and decision trees (Belgiu and Drȃgu, 2016). Fundamentally, it is an ensemble learning method that is efficient, simple to parametrize, robust, and often used for crop type and cropping pattern mapping (Feyisa et al., 2020;Liu and Chen, 2019;Richard et al., 2017). Based on the spectral analysis and VIs, five random classification scenarios were done viz. ...
Intercropping-the planting of more than one crop in the same plot of land-is a prevalent agricultural management practice which can be used for risk reduction. Despite its widespread use, intercropping is not commonly reported in agricultural statistics, resulting to very limited spatially disaggregated information about its prevalence. Remote sensing-based approaches to detect and estimate the area of cropping patterns like intercropping require good understanding of the spectral response of (intercropped) crops at different crop growth phases. This study integrates field surveys, farmer interviews and temporal Sentinel-2 data from four crop growth phases and the post-harvest period of maize and intercropped maize (imaize). The goal is to identify the optimal crop growth phases, spectral regions and vegetation indices (VIs) that can accurately discriminate the two cropping patterns. We computed p-values for the spectral bands using Mann-Whitney U test and identified critical crop growth phases. Classification of maize and imaize cropping patterns was performed using random forest classifier. Our spectral analysis revealed effective discrimination between maize and imaize cropping patterns during the vegetative (in all spectral bands) and flowering-yield phases (in Blue, Green, Red, RE704, RE783, NIR833, NIR865). The most suitable VIs contained red-edge and near-infrared spectal bands. Utilizing spectral data and VIs from vegetative and flowering-yield phases, we achieved optimal discrimination during the vegetative phase (user's accuracy of 100 % and producer's accuracy of 100 %). However, accuracy decreased during the flowering yield phase (overall accuracy of 87 % for all spectral bands). The highest classification results using all spectral bands at the flowering yield phase resulted in 80 % producer's accuracy for maize and 100 % for imaize. This study illustrates the utility of temporal Sentinel-2 spectral data for identifying the critical crop growth phase, spectral regions and VIs for cropping patterns classification, particularly for intercropping.
... Liu et al. 2022) while cropping patterns contain temporal crop planting sequences (Waldhoff, Lussem, and Bareth 2017) and spatial arrangements (P. Liu and Chen 2019). On the other hand, the information about crop type refers to specific crop varieties (Dahal, Wylie, and Howard 2018), while water management practices include irrigation availability (Qian et al. 2022) and the source of irrigation (Feyisa et al. 2020). ...
Accurate acquisition of spatial and temporal distribution information for cropping systems is important for agricultural production and food security. The challenges of extracting information about cropping systems in regions with smallholder farms are considerable, given the varied crops, complex cropping patterns, and the fragmentation of cropland with frequent reclamation and abandonment. This study presents a specialized workflow to solve this problem for regions with smallholder farms, which utilizes field samples and Sentinel-2 data to extract cropping system information over multiple years. The workflow involves four steps: 1) processing Sentinel-2 data to simulate crop growth curves with the Savitzky‒Golay filter and computing feature variables for classification, including phenology indices, spectral bands, and time series of vegetation indices; 2) mapping annual croplands with one-class support vector machine; 3) mapping various cropping patterns, including single cropping, intercropping, double cropping, multiple harvest, and fallow by decision tree and K-means clustering; and 4) mapping crops with random forest where Jeffries-Matusita distance was used to select appropriate vegetation indices. The workflow was applied in the Hetao irrigation district in Inner Mongolia Autonomous Region, China from 2018 to 2021. The overall accuracies were 0.98, 0.96, and 0.97 for cropland, cropping patterns, and crop type mapping, respectively. The mapping results indicated that the study area has low cropping continuity and is dominated by single cropping patterns. Furthermore, the area of wheat cultivation has decreased, and vegetable cultivation has expanded. Overall, the proposed workflow facilitated the accurate acquisition of cropping system information in regions with smallholder farms and demonstrated the effectiveness of available Sentinel-2 imagery in classifying complex cropping patterns. The workflow is available on Google Earth Engine.
... This method has been widely used in remote sensing image classifications, target object detections, scene recognitions, and other tasks and has gradually become a means by which the semantic segmentation of remote sensing images can be performed [42,43]. Compared to traditional object-oriented and pixel-oriented classification methods [44][45][46], CNN-based networks typically achieve higher classification accuracy. For example, SegNet, FCN8s, and U-Net utilize automatic feature learning to avoid complex feature design, improving the automation and intelligence of remote sensing image segmentation [47]. ...
At present, forest and fruit resource surveys are mainly based on ground surveys, and the information technology of the characteristic forest and fruit industries is evidently lagging. The automatic extraction of fruit tree information from massive remote sensing data is critical for the healthy development of the forest and fruit industries. However, the complex spatial information and weak spectral information contained in high-resolution images make it difficult to classify fruit trees. In recent years, fully convolutional neural networks (FCNs) have been shown to perform well in the semantic segmentation of remote sensing images because of their end-to-end network structures. In this paper, an end-to-end network model, Multi-Unet, was constructed. As an improved version of the U-Net network structure, this structure adopted multiscale convolution kernels to learn spatial semantic information under different receptive fields. In addition, the “spatial-channel” attention guidance module was introduced to fuse low-level and high-level features to reduce unnecessary semantic features and refine the classification results. The proposed model was tested in a characteristic high-resolution pear tree dataset constructed through field annotation work. The results show that Multi-Unet was the best performer among all models, with classification accuracy, recall, F1, and kappa coefficient of 88.95%, 89.57%, 89.26%, and 88.74%, respectively. This study provides important practical significance for the sustainable development of the characteristic forest fruit industry.
... One major application of GF-1 images is in agricultural monitoring. Two sensors are carried on the GF-1 satellites, including wide field view (WFV) and panchromatic multispectral sensors (PMSs) cameras [23,24]. Zhou et al. [25], Chen et al. [26], and Qu et al. [27] used the GF-1 WFV image with 16 m spatial resolution to monitor crop lodging. ...
Efficient, fast, and accurate crop lodging monitoring is urgent for farmers, agronomists, insurance loss adjusters, and policymakers. This study aims to explore the potential of Chinese GF-1 PMS high-spatial-resolution images for corn lodging monitoring and to find a robust and efficient way to identify corn lodging accurately and efficiently. Three groups of image features and five machine-learning approaches are used for classifying non-lodged, moderately lodged, and severely lodged areas. Our results reveal that (1) the combination of spectral bands, optimized vegetation indexes, and texture features classify corn lodging with an overall accuracy of 93.81% and a Kappa coefficient of 0.91. (2) The random forest is an efficient, robust, and easy classifier to identify corn lodging with the F1-score of 0.95, 0.92, and 0.95 for non-lodged, moderately lodged, and severely lodged areas, respectively. (3) The GF-1 PMS image has great potential for identifying corn lodging on a regional scale.
... It must be noted that different crops might have different phenologic states characterized by different NDVI values. Moreover, the crops can have different basal NDVI values [35][36][37]. These basal differences can cause misclassifications of the crop with regular vigour as a crop with low vigour leading to unneeded treatment. ...
Mixed crops are one of the fundamental pillars of agroecological practices. Row intercropping is one of the mixed cropping options based on the combination of two or more species to reduce their impacts. Nonetheless, from a monitoring perspective, the coexistence of different species with different characteristics complicates some processes, requiring a series of adaptations. This article presents the initial development of a procedure that differentiates between chickpea, lentil, and ervil in an intercropping agroecosystem. The images have been taken with a drone at the height of 12 and 16 m and include the three crops in the same photograph. The Vegetation Index and Soil Index are used and combined. After generating the index, aggregation techniques are used to minimize false positives and false negatives. Our results indicate that it is possible to differentiate between the three crops, with the difference between the chickpea and the other two legume species clearer than that between the lentil and the ervil in images gathered at 16 m. The accuracy of the proposed methodology is 95% for chickpea recognition, 86% for lentils, and 60% for ervil. This methodology can be adapted to be applied in other crop combinations to improve the detection of abnormal plant vigour in intercropping agroecosystems.
... The GF-1 satellite has a short revisit cycle of 4 days and a wide scanning swath of 800 km, which is important for observing the lodging phenomenon since lodging usually occurs instantly and over a large area. Table 1 provides an overview of the relevant GF-1 data [26]. Two images for Zhaodong City and one for Ningjiang District retrieved on 6 September 2020 were used to study the spectral characteristics of lodged maize and distinguish lodged maize from non-lodged maize in this paper. ...
... Several studies were conducted on maize extraction using optical satellite data, and relatively mature extraction methods were developed [29,30]. The SVM method, a non-parametric algorithm, is widely used to perform maize mapping with a high accuracy [26,[31][32][33]. Considering the crop calendar and avoiding the lodging effect, the images taken before the lodging event were used to map the spatial distribution of maize using the SVM classifier. ...
Crop lodging is a major destructive factor for agricultural production. Developing a cost-efficient and accurate method to assess crop lodging is crucial for informing crop management decisions and reducing lodging losses. Satellite remote sensing can provide continuous data on a large scale; however, its utility in detecting lodging crops is limited due to the complexity of lodging events and the unavailability of high spatial and temporal resolution data. Gaofen1 satellite was launched in 2013. The short revisit cycle and wide orbit coverage of the Gaofen1 satellite make it suitable for lodging identification. However, few studies have explored lodging detection using Gaofen1 data, and the operational application of existing approaches over large spatial extents seems to be unrealistic. In this paper, we discuss the identification method of lodged maize and explore the potential of using Gaofen1 data. An analysis of the spectral features after maize lodging revealed that reflectance increased significantly in all bands, compared to non-lodged maize. A spectral sum index was proposed to distinguish lodged and non-lodged maize. Two study areas were considered: Zhaodong City in Heilongjiang Province and Ningjiang District in Jilin Province. The results of the identified lodged maize from the Gaofen1 data were validated based on three methods: first, ground sample points exhibited the overall accuracies of 92.86% and 88.24% for Zhaodong City and Ningjiang District, respectively; second, the cross-comparison differences of 1.01% for Zhaodong City and 1.13% for Ningjiang District were obtained, compared to the results acquired from the finer-resolution Planet data; and third, the identified results from Gaofen1 data and those from farmer survey questionnaires were found to be consistent. The validation results indicate that the proposed index is promising, and the Gaofen1 data have the potential for rapid lodging monitoring.
... Inclusion of Sentinel-1 has the main advantage that it is not restricted by cloud coverage, a common barrier during the tropical rainy season. There are a growing number of studies integrating these two data types in plantation mapping (Liu and Chen, 2019;Poortinga et al., 2019), annual crop mapping (Clerici et al., 2017;Denize et al., 2018;Jin et al., 2019;Mercier et al., 2019;Qadir and Mondal, 2020;Sun et al., 2019) and even detection of single trees outside of forests (Brandt and Stolle, 2021); however, the integration of Sentinel-1 and 2 data has not yet been applied to coffee production systems. ...
... While our model targets coffee, we drew from mapping literature on similar tree crops: shade cocoa, coconut, rubber, and oil palm. While separating tree crop from surrounding land cover such as forest and other agriculture land, we would emphasize, in continuation of previous work, the added value of texture features in distinguishing landscape and cropping systems (Burnett et al., 2019;Gao et al., 2015;Gomez et al., 2010;Liu and Chen, 2019;Numbisi et al., 2019). We would also make the distinction between tree crops with a regular clearing rotation (such as rubber) and fruit and nut trees (such as a coffee and cocoa), and between monoculture dominated landscape, agroforestry dominated landscapes, and mixed monoculture and agroforestry. ...
Perennial commodity crops, such as coffee, often play a large role globally in agricultural markets and supply chains and locally in livelihoods, poverty reduction, and biodiversity. Yet, the production of spatial information on these crops are often overlooked in favor of annual food crops. Remote sensing detection of coffee faces a particular set of challenges due to persistent cloud cover in the tropical “coffee belt,” hilly topography in coffee growing regions, diversity of coffee growing systems, and spectral similarity to other tree crops and agricultural land. Looking at the major coffee growing region in Dak Lak, Vietnam, we integrate multi-temporal 10 m optical Sentinel-2 and Sentinel-1 SAR data in order to map three coffee production systems: i) open-canopy sun coffee, ii) intercropped and other shaded coffee and iii) newly planted or young coffee.
Leveraging Google Earth Engine (GEE), we compute five sets of features in order to best enhance separability between coffee and other land cover and within coffee production systems. The features include Sentinel-2 dry and wet season composites, Sentinel-1 texture features, Sentinel-1 spatiotemporal metrics, and topographic features. Using a random forest classification algorithm, we produce a 9-class land cover map including our three coffee production classes and a binary coffee/non-coffee map. The binary map has an overall accuracy of 89% and the three coffee production systems have user accuracies of 65, 56, 71% for sun coffee, intercropped coffee and newly planted coffee, respectively. This is a first effort at large-scale distinction of within-crop production styles and has implications across many applications. The binary coffee map can be used as a high-resolution crop mask, whereas the detailed land cover map can inform monitoring of deforestation dynamics, biodiversity, sustainability certification and implementation of climate adaptation strategies. This work offers a scalable approach to integrating optical and radar Sentinel data for production of spatially explicit agricultural information and contributes particularly to tree crop and agroforestry mapping, which often is overlooked in between agricultural and forestry sciences.
... Liu et al. (2018) demonstrated that GF-2 performed better in detecting patchy vegetation than CBERS-04 (China-Brazil Earth Resources Satellite 4) and GF-1 (Gaofen-1) [20]. Liu et al. (2019) found that the textural, spectral, and multitemporal features extracted from GF-2 images could provide plenty information on intercropping regions [21]. Nonetheless, some issues still exist in mapping mangrove forests with GF-2 images, such as the limited spectral bands, the complex pre-processing procedures, and costly price, which are common problems in very high-resolution satellite images. ...
... Liu et al. (2018) demonstrated that GF-2 performed better in detecting patchy vegetation than CBERS-04 (China-Brazil Earth Resources Satellite 4) and GF-1 (Gaofen-1) [20]. Liu et al. (2019) found that the textural, spectral, and multitemporal features extracted from GF-2 images could provide plenty information on intercropping regions [21]. Nonetheless, some issues still exist in mapping mangrove forests with GF-2 images, such as the limited spectral bands, the complex pre-processing procedures, and costly price, which are common problems in very high-resolution satellite images. ...
Mangrove forest extents and distributions are fundamental for conservation and restoration efforts. According to previous studies, both the commercial Gaofen-2 (GF-2) imagery (0.8 m spatial resolution and 4 spectral bands) and freely accessed Sentinel-2 (S2) imagery (10 m spatial resolution and 13 spectral bands) have been successfully used to map mangrove forests. However, the efficiency and accuracy of mangrove forest mapping based on these two data is not clear, especially for large-scale applications. To address this issue, firstly, we developed a robust classification approach by integrating object-based image analysis (OBIA) and random forest (RF) algorithm; and then, applied this approach to GF-2 and S2 images to map the extents of mangrove forest along the entire coasts of Guangxi, China, respectively; at last, compared the efficiency and accuracy of GF-2 and S2 imagery in mangrove forest mapping. Results showed that: (1) based on OBIA and RF integrated classification approach both mangrove forest maps derived from GF-2 and S2 obtained high mapping accuracies (the overall accuracy was 96% and 94%, respectively); (2) areal extent of mangrove forests in Guangxi extracted from GF-2 and S2 images was 8182 ha and 8040 ha, respectively; (3) GF-2 imagery has extraordinary abilities in detecting fragmented mangrove forest patches located along landward and seaward edges; (4) S2 imagery performed better in detecting seaward submerged mangrove forests and separating mangrove forest from terrestrial vegetation. Results and conclusions of this study can provide basic considerations for selecting appropriate data source in mangrove forest or wetland vegetation mapping tasks.
... 2.3.1. SVM Support vector machine (SVM) is a type of generalized linear classifier for binary classification by supervised learning, and has been widely used for remote sensing [37,40] and computer graphics [41,42]. It can find an optimal hyperplane in the feature space and divide the data into two categories ( Figure 8). ...
... Support vector machine (SVM) is a type of generalized linear classifier for binary classification by supervised learning, and has been widely used for remote sensing [37,40] and computer graphics [41,42]. It can find an optimal hyperplane in the feature space and divide the data into two categories ( Figure 8). ...
In the case of marine accidents, monitoring marine oil spills can provide an important basis for identifying liabilities and assessing the damage. Shipborne radar can ensure large-scale, real-time monitoring, in all weather, with high-resolution. It therefore has the potential for broad applications in oil spill monitoring. Considering the original gray-scale image from the shipborne radar acquired in the case of the Dalian 7.16 oil spill accident, a complete oil spill detection method is proposed. Firstly, the co-frequency interferences and speckles in the original image are eliminated by preprocessing. Secondly, the wave information is classified using a support vector machine (SVM), and the effective wave monitoring area is generated according to the gray distribution matrix. Finally, oil spills are detected by a local adaptive threshold and displayed on an electronic chart based on geographic information system (GIS). The results show that the SVM can extract the effective wave information from the original shipborne radar image, and the local adaptive threshold method has strong applicability for oil film segmentation. This method can provide a technical basis for real-time cleaning and liability determination in oil spill accidents.
The largest green tide in the South Yellow Sea (SYS) of China broke out in 2021 since the first outbreak in 2007. What causes the great outbreak has caused widespread concern. In this study, we monitor the green tide in the SYS in 2021 using Normalized Difference Vegetation Index and Google Earth Engine and then identify the main controlling factors from environmental change and human activity. In terms of human activity, Pyropia yezoensis (P. yezoensis) cultivation in Northern Jiangsu shoal of China has an impact on the scale of green tide. Compared with the same period in previous years, the early recycling of P. yezoensis cultivation rafts in 2021 provided more floating seed for green tide. The human activity intensity index in coastal cities in Jiangsu and Shandong provinces of China has been increasing for the past nine years, leading to large amounts of nutrients being released into seawater, providing green tide with the nutrients needed for growth. From the perspective of environmental changes, the sea surface temperature was higher in June and July in the SYS, which was favorable for green tide growth. The abundant solar radiation in May and July is conducive to photosynthesis of Ulva prolifera. The lower photosynthetically active radiation in August is conducive to reduce the effect of photoinhibition and promote algal growth. These findings are useful for further understanding the rules of green tide outbreaks and controlling green tide disasters.
