Satellite imagery contains valuable large-scale information for precision farming. However, the low-resolution of satellite images can make it challenging to extract crop status information due to mixed pixels, in particular within multi-species crop stands like grass-clover for silage. In contrast, proximal high-resolution images with centimeter to sub-millimeter scale contain cures for single crop species pixels. However, these are often sparsely sampled due to computational limitations. In this paper, we present a preliminary attempt to enrich multispectral satellite images with crop stand population intelligence extracted from sparsely proximal RGB samples. The system attempts to reinforce satellite imagery based on proximal indicators lowering the risk of faulty interpretation knowledge base for future farm management information system (FMIS). A semantic segmentation algorithm is utilized to find the ratio of grass, clover, and soil across proximal images. Sentinel-2 as satellite imagery is employed as the 10-meter ground sampling distance input of the system and the grass, clover, and soil ratios are the output gained simultaneously. The system includes 1) a method where the proximal images and satellite imagery are preprocessed and then aligned with each other; and 2) a non-linear Multi-Layer Perceptron (MLP) extracting grass, clover, and soil ratio. Estimation results present promising correlation between clover, grass, soil, and Sentinel-2. Although, more data with higher diversity of clover-grass mixture is required to confirm the distinction of clover and grass.
KEYWORDS: Video, Video processing, Digital filtering, Denoising, Image filtering, Analog electronics, LCDs, Televisions, Principal component analysis, Image quality standards
In this paper we present a noise reduction filter for video processing. It is based on the recently proposed two
dimensional steering kernel, extended to three dimensions and further augmented to suit the spatial-temporal
domain of video processing. Two alternative fillters are proposed - the time symmetric kernel and the time
asymmetric kernel. The first reduces the noise on single sequences, but to handle the problems at scene shift
the asymmetric kernel is introduced. The performance of both kernels is tested on simulated data and on a
real video sequence together with the original steering kernel. The proposed kernels improves the Rooted Mean
Squared Error (RMSE) compared to the original steering kernel method on video material significantly.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.