Remote sensing has been increasingly used in precision agriculture. Buoyed by the developments in the miniaturization of sensors and platforms, contemporary remote sensing offers data at resolutions finer enough to respond to within-farm variations. LiDAR point cloud, offers features amenable to modelling structural parameters of crops. Early prediction of crop growth parameters helps farmers and other stakeholders dynamically manage farming activities. The objective of this work is the development and application of a deep learning framework to predict plant-level crop height and crown area at different growth stages for vegetable crops. LiDAR point clouds were acquired using a terrestrial laser scanner on five dates during the growth cycles of tomato, eggplant and cabbage on the experimental research farms of the University of Agricultural Sciences, Bengaluru, India. We implemented a hybrid deep learning framework combining distinct features of long-term short memory (LSTM) and Gated Recurrent Unit (GRU) for the predictions of plant height and crown area. The predictions are validated with reference ground truth measurements. These predictions were validated against ground truth measurements. The findings demonstrate that plant-level structural parameters can be predicted well ahead of crop growth stages with around 80% accuracy. Notably, the LSTM and the GRU models exhibited limitations in capturing variations in structural parameters. Conversely, the hybrid model offered significantly improved predictions, particularly for crown area, with error rates for height prediction ranging from 5 to 12%, with deviations exhibiting a more balanced distribution between overestimation and underestimation This approach effectively captured the inherent temporal growth pattern of the crops, highlighting the potential of deep learning for precision agriculture applications. However, the prediction quality is relatively low at the advanced growth stage, closer to the harvest. In contrast, the prediction quality is stable across the three different crops. The results indicate the presence of a robust relationship between the features of the LiDAR point cloud and the auto-feature map of the deep learning methods adapted for plant-level crop structural characterization. This approach effectively captured the inherent temporal growth pattern of the crops, highlighting the potential of deep learning for precision agriculture applications.
30 November 2021 in ISPRS Journal of Photogrammetry and Remote Sensing
Atmospheric correction is an important pre-processing step in various spatio-temporal and multi-sensor remote sensing data analyses based applications. Absolute atmospheric corrections are carried out using physical based models, generally known as radiative transfer (RT) codes such as MODerate resolution atmospheric TRANsmission (MODTRAN), Second Simulation of the Satellite Signal in the Solar Spectrum (6S) etc. Most of the available atmospheric correction schemes are commercially off-the-shelf (COTS) and use patented RT codes. The objective of the present work is to develop an open-end atmospheric correction scheme, named as Flexible Atmospheric Compensation Technique (FACT), based on open source 6S RT code. The proposed FACT scheme utilizes a look-up architecture for simulating the outputs of 6S RT code for various input parameters’ combination. Input parameters such as initial visibility, columnar water vapour are estimated using the dark object and the Continuum Interpolated Band Ratio (CIBR) methods respectively. The proposed FACT scheme has been evaluated exhaustively using spatio-spectral statistical error measures such as Spatial-Root Mean Square Error (S-RMSE), Spatial-Mean Absolute Error (S-MAE), and spectral-RMSE by comparing the performance with widely used Fast Line-of-sight Atmospheric Analysis of Hypercubes (FLAASH), and the customized NASA JPL’s atmospheric correction scheme. Datasets from hyperspectral (AVIRIS-NG and Hyperion) and multispectral (LANDSAT-8 OLI and WorldView-3) remote sensors were chosen for comparative analysis of the developed atmospheric correction scheme against other atmospheric correction schemes. Results confirm that the proposed FACT scheme offers accuracy of about 95% for hyperspectral imaging sensors and close to 98% for multispectral imaging sensors when compared with FLAASH. Despite marginal disagreements for certain land cover features at the water vapour absorbing spectral regions, we find the proposed FACT scheme a plausible option for carrying out absolute atmospheric correction of various operational remote imaging sensors.