Search Results (2)

To improve the multi-resolution segmentation (MRS) quality of plastic greenhouses (PGs) in GaoFen-2 (GF-2) images, the effects of atmospheric correction and image enhancement on effective PG segments (EPGSs) were evaluated. A new semi-automatic method was also proposed to extract EPGSs in an accurate and efficient way. Firstly, GF-2 images were preprocessed via atmospheric correction, orthographical correction, registration, fusion, linear compression, or spatial filtering, and, then, boundary-removed point samples with adjustable density were made based on reference polygons by taking advantage of the characteristics of chessboard segmentation. Subsequently, the point samples were used to quickly and accurately extract segments containing 70% or greater of PG pixels in each MRS result. Finally, the extracted EPGSs were compared and analyzed via intersection over union (IoU), over-segmentation index (OSI), under-segmentation index (USI), error index of total area (ETA), and composite error index (CEI). The experimental results show that, along with the change in control variables, the optimal scale parameter, time of segmentation, IoU, OSI, USI, and CEI all showed strong changing trends, with the values of ETA all close to 0. Furthermore, compared with the control group, all the CEIs of the EPGSs extracted from those corrected and enhanced images resulted in lower values, and an optimal CEI involved linearly compressing the DN value of the atmospheric-corrected fusion image to 0–255, and then using Fast Fourier Transform and a circular low-pass filter with a radius of 800 pixels to filter from the spatial frequency domain; in this case, the CEI had a minimum value of 0.159. The results of this study indicate that the 70% design in the experiment is a reasonable pixel ratio to determine the EPGSs, and the OSI-USI-ETA-CEI pattern can be more effective than IoU when it is needed to evaluate the quality of EPGSs. Moreover, taking into consideration heterogeneity and target characteristics, atmospheric correction and image enhancement prior to MRS can improve the quality of EPGSs. Full article

Compared to multispectral or panchromatic bands, fusion imagery contains both the spectral content of the former and the spatial resolution of the latter. Even though the Estimation of Scale Parameter (ESP), the ESP 2 tool, and some segmentation evaluation methods have been introduced to simplify the choice of scale parameter (SP), shape, and compactness, many challenges remain, including obtaining the natural border of plastic greenhouses (PGs) from a GaoFen-2 (GF-2) fusion imagery, accelerating the progress of follow-up texture analysis, and accurately evaluating over-segmentation and under-segmentation of PG segments in geographic object-based image analysis. Considering the features of high-resolution images, the heterogeneity of fusion imagery was compressed using texture analysis before calculating the optimal scale parameter in ESP 2 in this study. As a result, we quantified the effects of image texture analysis, including increasing averaging operator size (AOS) and decreasing greyscale quantization level (GQL) on PG segments via recognition of a proposed Over-Segmentation Index (OSI)-Under-Segmentation Index (USI)-Error Index of Total Area (ETA)-Composite Error Index (CEI) pattern. The proposed pattern can be used to reasonably evaluate the quality of PG segments obtained from GF-2 fusion imagery and its derivative images, showing that appropriate texture analysis can effectively change the heterogeneity of a fusion image for better segmentation. The optimum setup of GQL and AOS are determined by comparing CEI and visual analysis. Full article