Search

Ultra-high-resolution aerial images obtained from Unmanned Aerial Vehicles (UAVs) have widely been used to extract crop planting information in recent years. However, some high-resolution multispectral or hyperspectral images were usually costly and time-consuming for data processing. Therefore, it is very necessary to effectively use easily accessible and low-cost high-resolution RGB images, particularly to eliminate the common noises (e.g., shadows and bare land) for a better extraction accuracy of crop planting. In this study, a high-precision extraction method was proposed to obtain the maize planting information using 1.8 cm resolution UAV aerial orthophotos (i.e., RGB images). The experimental maize farm was located in Southeast Africa, where images were taken at noon during the maize growing season. The classification features were also selected from the aspects of the spectrum, color space, and image texture. Then, five types of classification were selected to extract maize planting information, including Bayes, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) algorithms). Firstly, an object-oriented interpreting platform (eCognition9.0 software) was selected to calculate the space transform of Hue, Saturation, and Intensity (HSI) color, eight types of RGB texture, and five vegetation indices, including the normalized green-red difference index, red-green ratio index, vegetation color verification index, visible-band difference vegetation index, and excess green vegetation index. Then, three types of feature space were constructed: 1) The first feature space was composed of three sub-feature spaces, i.e., vegetation indices, HSI color space features, and RGB image texture features; 2) The second feature space was composed of four sub-feature spaces, where three types of features were combined (i.e., vegetation indices, HSI color space, and RGB image texture) in pairs or total; 3) The third feature space was composed of the most optimal factors, where the dimension reduction was performed on the combination of all three types of features using RF. Subsequently, the RGB images were classified into three land-use types, including maize, bare land, and shadow. Bayes, KNN, SVM, DT, and RFs models were finally selected for the supervised classification with error matrix. The results showed that the optimal classification accuracy was obtained using neither a single feature nor all three types of features in total. More importantly, a combination of features was usually achieved higher accuracy than that of a single feature. Specifically, the best choice was the combination of HSI color and RGB image texture features using the RF, particularly with the total highest accuracy of 0.862 and a Kappa coefficient of 0.793. Additionally, the dimension reduction of features using RF models was neither significantly improved nor reduced classification accuracy (except for the SVM). However, the factors retained from the feature dimension reduction were easily explained suitable for the actual background and meaning. Furthermore, both classification efficiency and stability were improved greatly during this time. The finding can provide a specific solution for the high-precision extraction of crop planting information using UAV RGB images [ABSTRACT FROM AUTHOR]

The wide use of plastic greenhouses in the word has brought economic benefits, but also caused many environmental problems. Accurate and timely acquisition of spatial distribution information of plastic greenhouse is of great significance to agricultural production and soil management. The use of plastic greenhouse has changed the structure of soil, thus changing the surface spectral characteristics of the soil. Because of the particularity of the plastic film material, the reflection spectrum has strong directionality and uncertainty, so it is difficult to identify plastic greenhouse accurately only depending on the reflectance spectrum characteristics of ground objects. In this paper, Kunming City, Yunnan Province is taken as the research area, using GF-2 image as a single data source, the multi-scale segmentation method is used to obtain the image object efficiently and accurately. According to the detailed spatial information of plastic greenhouse, the applicability of three image object-based texture extraction algorithms, namely GLCM(gray-level co-occurrence matrix), LBP(local binary pattern) and PSI(pixel shape index) for plastic greenhouse identification is analysed and tested. In addition, different texture features are combined with spectral features and NDVI index to form different classification schemes, to explore which one would be the best combination of texture features for identification of plastic greenhouses. In order to explore the robustness of the method, different texture feature combination schemes are applied in the study areas. The results of SVM (support vector machine) classifier are evaluated by confusion matrix. The results show that the overall combination schemes of the two research areas with different landscape patterns have the same trend. The combination of spectral features and NDVI index can accurately identify the scattered waters in each of the two study areas. For plastic greenhouses and impervious surfaces with similar reflectance spectra, adding texture features can make up for limitation of spectral characteristics and improve the overall accuracy. The phenomena of homologous or homologous spectra in spectral features of high spatial resolution image can effectively improve the discrimination between plastic greenhouse and impervious surface. Texture features can significantly improve the identification accuracy of plastic greenhouse following the object-based image classification frame. In the classification scheme of plastic greenhouses with different spatial distribution structures, the LBP (local binary pattern) texture algorithm has the best recognition accuracy, the overall accuracy of study area A is 96.85%, Kappa coefficient is 0.95, and that of study area B is 95.58% and 0.94. Landscape fragment analysis (landscape fragmentation index area mean index, aggregation indices) of the two different study areas showed that the plastic greenhouses in study area A are more fragmented than study area B (area mean indices are 3.39 hm² and 1.37 hm², aggregation indices are 80.64 and 72.98 for plastic greenhouses in study area A and B, respectively). The results of fragmentation are consistent with those of classification, and the accuracy of landscape classification with more space fragments is lower (the highest overall classification accuracy for study area A and B are 98.13% and 96.13%, respectively, the PA(producer accuracy) and UA(user accuracy) are 96.47% and 97.93% for study area A, and 90.67%, 99.68% for study area B). The results show that texture features based on image objects can improve the recognition accuracy of small-scale plastic greenhouse. This is of great significance to the accurate mapping of the distribution of plastic greenhouses. [ABSTRACT FROM AUTHOR]