Project

LESS: Efficient 3D Radiative Transfer Model

Goal: The projects aims to develop an efficient ray-tracing based 3D radiative transfer model to simulate various remote sensing datasets. For using this model, please refer to: http://lessrt.org/

Updates
0 new
8
Recommendations
0 new
26
Followers
0 new
48
Reads
11 new
394

Project log

Jianbo Qi
added 4 research items
Forest canopy cover (FCC) plays an important role in many ecological, hydrological and forestry applications. For large-scale applications, FCC is usually estimated from remotely sensed data by inverting radiative transfer models (RTMs) or using data-driven regressions. In this study, we proposed a hybrid model, which combines 3D RTMs and transfer learning-based convolutional neural network (T-CNN), to estimate FCC from very high-resolution satellite images (e.g., Chinese GaoFen-2, 1 m resolution with 4 bands). Unlike common hybrid models that are purely trained with simulation data, T-CNN combines simulation data-based pre-training and actual data-based transfer learning, which is a widely used technique in artificial intelligence for fine-tuning models. The performance of T-CNN was compared with a random forest (RF) model and two general CNN models, including CNN trained with actual dataset only (Data-CNN) and CNN trained with RTM simulation data only (RTM-CNN). Results on the independent validation dataset (not used in training stage) showed that T-CNN had higher accuracy (RMSE = 0.121, R2 = 0.83), compared with RF (RMSE = 0.26, R2 = 0.61), Data-CNN (RMSE = 0.142, R2 = 0.81), and RTM-CNN (RMSE = 0.144, R2 = 0.73), which indicates that T-CNN has a strong transferability. Tests on different training sizes showed that T-CNN (0.084<RMSE<0.108) provided constantly better performances than RF (0.116<RMSE<0.122) and Data-CNN (0.103<RMSE<0.128), which demonstrates the potential of T-CNN as an alternative to RTM-based inversion and data-driven regressions to estimate FCC, especially when training data is imbalanced and inadequate.
Optical remote sensing allows to efficiently monitor forest ecosystems at regional and global scales. However, most of the widely used optical forward models and backward estimation methods are only suitable for forest canopies in flat areas. To evaluate the recent progress in forest remote sensing over complex terrain, a satellite-airborne-ground synchronous Fine scale Optical Remote sensing Experiment of mixed Stand over complex Terrain (FOREST) was conducted over a 1 km×1 km key experiment area (KEA) located in the Genhe Reserve Areain 2016. Twenty 30 m×30 m elementary sampling units (ESUs) were established to represent the spatiotemporal variations of the KEA. Structural and spectral parameters were simultaneously measured for each ESU. As a case study, we first built two 3D scenes of the KEA with individual-tree and voxel-based approaches, and then simulated the canopy reflectance using the LargE-Scale remote sensing data and image Simulation framework over heterogeneous 3D scenes (LESS). The correlation coefficient between the LESS-simulated reflectance and the airborne-measured reflectance reaches 0.68–0.73 in the red band and 0.56–0.59 in the near-infrared band, indicating a good quality of the experiment dataset. More validation studies of the related forward models and retrieval methods will be done.
Three-dimensional (3D) radiative transfer simulations are critical for studying the radiometric properties of canopies. Efficient and easy-to-use 3D radiative transfer models are required by remote sensing inversion and many validation applications. Extensive efforts have been made to improve the computational efficiency, accuracy, and useability of 3D radiative transfer models. This study focuses on the abstraction of canopies for 3D radiative transfer simulations by proposing a lightweight boundary-based description of leaf clusters (B-cluster) to ease the creation of 3D scenes while keeping the simulation as accurate as possible. B-cluster partitions a tree crown into sub-crown leaf clusters and abstracts each of them into a turbid medium enclosed by a complex and tight boundary, while terrain and branches are described with precise mesh surfaces. The radiative transfer simulation within B-cluster has been developed based on an efficient Monte Carlo path-tracing algorithm and implemented in the LargE-Scale remote sensing data and image Simulation framework (LESS) model by considering the presence of both turbid medium and surface scattering. The performance of the model was assessed by comparing with original LESS version, which describes all landscape elements with mesh surfaces (here called M-surface approach), and with a uniform voxel-based approach (U-voxel) in terms of the multiangle bidirectional reflectance factor (BRF) as well as with pixel-wise images. Results show that B-cluster is highly consistent with M-surface in abstract canopies (mean normalized absolute BRF differences δ¯ < 2%) and in realistic forest stand (δ¯ < 5% at 5-m resolution) with considerably reduced requirements for computational memory. Compared with U-voxel, B-cluster is also more robust and better at describing canopy structures with different levels of detail. B-cluster enables to quickly construct accurate 3D scenes with reduced requirements of computational resources. It is also a unified and scale-adaptive approach that can describe crowns as simple as geometric primitives and as complex as explicitly described meshes. The newly proposed approach has been released in new LESS versions at http://lessrt.org/.
Jianbo Qi
added a research item
Chlorophyll content is a key trait for understanding the functioning of agroforestry ecosystems and has important implications for leaf and canopy photosynthesis. However, fine-scale monitoring of canopy chlorophyll content (CCC) of individual fruit trees is rather challenging. This study aims to use a 3D radiative transfer model (RTM) and proposes a joint inversion model based on prior knowledge to estimate the CCC of individual tree crowns (ITCs) in apple orchards. The widely recognized 3D RTM LESS (large-scale remote sensing data and image simulation framework over heterogeneous 3D scenes) was adopted for large-scale apple orchard 3D scenes radiative transfer computing and image simulation. LESS was first evaluated with unmanned aerial vehicle (UAV) multispectral imagery and the results showed that it reasonably characterized the reflectance of apple tree canopies (RMSE = 0.02). An original look-up table (LUT) with reflectance was then produced using LESS, and the final vegetation indices LUT (VI LUT) including Normalized Difference Vegetation Index (NDVI), Green Chlorophyll Index (CIgreen), Red edge Chlorophyll Index (CIred edge) and Green NDVI (GNDVI) was generated from the original LUT form VI interpolation. A physically-based joint inversion model coupling prior knowledge of leaf pigments and leaf area index (LAI) was developed to estimate the CCC of ITCs from high-resolution UAV images. The solution first used linear interpolation to produce a weighted VI LUT corresponding to the sample based on estimated LAI. Linear interpolation was then adopted to screen multiple combinations of leaf chlorophylla+b (Cab) and leaf carotenoids (Cxc) contents from the VI LUT. A prior relationship between Cab and Cxc was finally used to regularize the constraints on multiple VI combinations and determine the estimation of Cab and CCC. The joint inversion model demonstrated an accurate estimation of CCC of ITCs. The model driven by GNDVI yielded the highest result for CCC estimation (R² = 0.84, RMSE = 24.12 μg/cm²). In addition, CIgreen (R² = 0.82, RMSE = 32.22 μg/cm²) and CIred edge (R² = 0.81, RMSE = 34.05 μg/cm²) also achieved satisfactory results. The proposed model facilitates CCC estimation of ITCs from high-resolution imagery in heterogeneous orchard canopies, which is important for advancing the precise nutrition management of fruit trees.
Jianbo Qi
added an update
Recently, we have introduced a new module in LESS, named LESS1D. It only requires a few statistical parameters, like LAI, LAD in some common 1D radiative transfer models, however, the radiation transfer process in LESS1D is fully computed with ray tracing, thus , it gives the same accurate results with 3D radiative trasnfer models. Currently, LESS1D supports homogeneous, row, and discrete canopies. Using LESS1D, it is more convenient to simulate large number of simulations with different parameter combinations, for generating look up tables or training ANNs. LESS1D is integrated in LESS, http://lessrt.org/
 
Jianbo Qi
added an update
In year of 2021, LESS has been used by a number of researches, it has been used to validate remote sensing physical models, perform parameter sensitivity analysis, as well as instruct field measurements. The list of papers that has used LESS can be seen from: http://lessrt.org/publications/
 
Jianbo Qi
added an update
Since the last release (almost one year ago), LESS has been enhanced in several aspects, we are pleased to release a new version, which includes:
  • An updated GUI based Batch Tool to support more parameters.
  • a complete new Python SDK, which enables users to control almost every parameters (including scene objects) for batch processing.
  • A new simple tree crown tool to create tree crowns by inputing paramters, such as leaf angle distribution.
  • Some bugs reported by users have been fixed (many thanks for their support).
  • A new LiDAR simulator tool (Extensions->Lidar Simulator) has been released.
This new version has been supported and contributed by many people, thanks for your contribution.
Please refer to http://lessrt.org/ to download this new version.
 
Jianbo Qi
added an update
LESS has been used to validate FPAR retrival algorithm, which shows the potentials of LESS to validate remote sensing products, especially for coarse-resolution products that are usually difficult to validate by using field measurements. In the near future, more applications will be shown, such clumping index, etc.
 
Jianbo Qi
added an update
A new version of LESS has been released. In this version:
1. Ubuntu 16.04 has been officially supported.
2. Some new features has been implemented, such as FPAR/Albedo simulation, output four-components image ( i.e., 1-illuminated soil, 2- illuminated object, 3-shaded soil, 4-shaded object. )
 
Jianbo Qi
added a research item
Accurate and efficient measurement of leaf angle distribution (LAD) is important for characterizing canopy structures and understanding solar radiation regimes within the plant canopy. The main challenge for obtaining LAD is measuring the orientations of individual leaves rapidly and accurately in complex field conditions. In this letter, we propose an efficient and low-cost approach to estimate both leaf zenith and azimuth angles from smartphone photographs by using a structure from motion (SfM) point cloud and pyramid convolutional neural network (PCNN)-based leaf detection. This SfM-PCNN method first detects individual leaves from 2-D photographs by delineating leaf boundaries, while minimizing the influences of interior leaf textures. The segmented image with leaf annotations is then used to partition the 3-D SfM point cloud into leaf clusters, each of which is fit by a plane to calculate the leaf orientation. The method was validated with manual measurements for five plant species with different leaf sizes, leaf shapes, and leaf textures. The accuracy is satisfactory for a leaf-to-leaf comparison over a Euonymus japonicus Thunb. with R-squared values of 0.84 (RMSE = 6.27°) and 0.97 (RMSE = 12.61°) for zenith and azimuth angle estimations, respectively. The method allows researchers to efficiently acquire LADs of different plants with low cost yet high accuracy.
Jianbo Qi
added an update
LESS has been used in some publications, such as :
1. Simulating images under different resolutions of crop field:
Li, L., Mu, X., Macfarlane, C., Song, W., Chen, J., Yan, K. and Yan, G., 2018. A half-Gaussian fitting method for estimating fractional vegetation cover of corn crops using unmanned aerial vehicle images. Agricultural and Forest Meteorology, 262, pp.379-390.
2. Simulating fisheye camera for validating sky view factor algorithms over urban areas:
Jiao, Z.-H., Ren, H., Mu, X., Zhao, J., Wang, T., Dong, J., n.d. Evaluation of Four Sky View Factor Algorithms using Digital Surface and Elevation Model Data. Earth and Space Science 0. https://doi.org/10.1029/2018EA000475
 
Jianbo Qi
added 5 research items
This paper presents a method to reconstruct individual trees from Terrestrial Laser Scanning (TLS) data obtained in leafoff conditions of an experiment plot. It firstly used the point clouds to build the branch structures of trees with a global optimization method. Computer generated needles and shoots were added to the previously constructed branches according to the leaf area (LA) of each individual tree, in consideration of clumping effect of small-scale structures. The LA was determined by the proportion of crown volume in this plot with LAI measured. In this way, several larix trees with different shapes and heights were reconstructed, which is basis of 3D forest scene reconstruction.
Three-dimensional (3D) radiative transfer modeling of the transport and interaction of radiation through earth surfaces is challenging due to the complexity of the landscapes as well as the intensive computational cost of 3D radiative transfer simulations. To reduce computation time, current models work with schematic landscapes or with small-scale realistic scenes. The computer graphics community provides the most accurate and efficient models (known as renderers) but they were not designed specifically for performing scientific radiative transfer simulations. In this study, we propose LESS, a new 3D radiative transfer modeling framework. LESS employs a weighted forward photon tracing method to simulate multispectral bidirectional reflectance factor (BRF) or flux-related data (e.g., downwelling radiation) and a backward path tracing method to generate sensor images (e.g., fisheye images) or large-scale (e.g. 1 km2) spectral images. The backward path tracing also has been extended to simulate thermal infrared radiation by using an on-the-fly computation of the sunlit and shaded scene components. This framework is achieved through the development of a user-friendly graphic user interface (GUI) and a set of tools to help construct the landscape and set parameters. The accuracy of LESS is evaluated with other models as well as field measurements in terms of directional BRFs and pixel-wise simulated image comparisons, which shows very good agreement. LESS has the potential in simulating datasets of realistically reconstructed landscapes. Such simulated datasets can be used as benchmarks for various applications in remote sensing, forestry investigation and photogrammetry.
Jianbo Qi
added a project goal
The projects aims to develop an efficient ray-tracing based 3D radiative transfer model to simulate various remote sensing datasets. For using this model, please refer to: http://lessrt.org/