Abstract
A workflow headed up to identify crops growing under plastic-covered greenhouses (PCG) and based on multi-temporal and multi-sensor satellite data is developed in this article. This workflow is made up of four steps: (i) data pre-processing, (ii) PCG segmentation, (iii) binary pre-classification between greenhouses and non-greenhouses, and (iv) classification of horticultural crops under greenhouses regarding two agronomic seasons (autumn and spring). The segmentation stage was carried out by applying a multi-resolution segmentation algorithm on the pre-processed WorldView-2 data. The free access AssesSeg command line tool was used to determine the more suitable multi-resolution algorithm parameters. Two decision tree models mainly based on the Plastic Greenhouse Index were developed to perform greenhouse/non-greenhouse binary classification from Landsat 8 and Sentinel-2A time series, attaining overall accuracies of 92.65% and 93.97%, respectively. With regards to the classification of crops under PCG, pepper in autumn, and melon and watermelon in spring provided the best results (Fβ around 84% and 95%, respectively). Data from the Sentinel-2A time series showed slightly better accuracies than those from Landsat 8.
Highlights
Growing needs in food supply will require higher agricultural yields [1]
The innovative goal faced in this paper relies on the integration of all the necessary steps to carry out the identification of under plastic-covered greenhouses (PCG) crops from using an object-based image analysis (OBIA) approach based on multi-temporal (Landsat 8 and Sentinel-2 time series) and multi-sensor (WorldView-2, Landsat 8, and Sentinel-2) satellite imagery. This workflow is composed of four stages: (i) data pre-processing; (ii) segmentation aimed at mapping greenhouse objects; (iii) binary pre-classification in GH or Non-GH; and (iv) classification of under PCG crops in two agronomic seasons
The scale parameter (SP), Shape, and Compactness parameters set to perform the MRS segmentation were 37.0, 0.4, and 0.5, respectively, yielding a modified Euclidean Distance 2 (ED2) value of 0.198 and a segmentation composed of 10,990 objects for the entire study area
Summary
Growing needs in food supply will require higher agricultural yields [1]. In this way, the use of plastic materials during the past 60 years, as a tool to move up the first harvest and increase the yield of horticultural crops, has been steadily increasing throughout the entire world [2]. In 2016, plastic-covered greenhouses (PCG) have reached a total coverage of 3019 million hectares over the world [3]. They are mainly localized in Europe, North Africa, the Middle East, and China [4]. The increasing role of agriculture in the management of sustainable natural resources calls for the development of operational cropland mapping and monitoring methodologies [5]. Due to their synoptic acquisitions and high revisit frequency, the data obtained by remote sensing can offer a significant contribution to provide periodic and accurate pictures of the agricultural sector [6,7]. It takes special relevance considering that a new era of land cover analysis has emerged, which has been enabled by free and open access data (e.g., Sentinel-2 or Landsat 8 images), analysis-ready data, high-performance computing, and rapidly developing data processing and analysis capabilities [8]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have