Traditional Pixel-based Approaches Research Articles

Geomorphological slope units of the Himalayas

ABSTRACT Slope units represent surface slopes by means of polygons delimited by drainage and divide lines obtained on a digital topography. Objective slope unit delineation for a given digital elevation model is still an open issue and, often, a limitation that may dictate the use of a more traditional pixel-based approach for spatial analysis. Availability of slope unit maps facilitates many kinds of studies and allows scholars to focus on specific scientific issues rather than on preparing sound mapping units from scratch for their research. Here, we present a slope unit map of a large portion of the Himalayas. The map is prepared following a widely tested, parameter-free optimization algorithm. The area encompassed by the map is relevant to studies of the well-known 2015 Gorkha earthquake and monsoons, which makes it relevant to a vast portion of the scientific community working in natural hazards including, but not limited to, landslide scientists and practitioners. The map contains 112,674 polygons with average area of 0.38 km and is published in vector form. The map is accompanied by a selection of data including morphometric and thematic quantities. In addition to describing the rationale behind the delineation of the polygonal map and selected data, we describe an application devoted to unsupervised terrain classification. We applied a k-means clustering procedure with two strategies: one at (coarser) basin scale and one at (finer) slope unit scale. We show similarities and differences between the two classification strategies, highlighting the role of the slope unit subdivision in the two cases.

A Novel Approach to Estimate Maize Lodging Area With PolSAR Data

Assessing crop lodging at the regional scale is an important requirement for breeding lodging-resistant varieties and harvest planning. Accurately and continuously estimating crop lodging area from remote sensing data remains challenging due to the high randomness scattering signal of SAR images and the insufficient number of applicable optical images. This study developed a new framework for estimating crop lodging area based on SAR data using the spatial aggregation approach of field units, overcoming the deficit of the traditional pixel-based approach susceptible to speckle noise and spatial heterogeneity. We aggregated the field’s pixel in SAR images using the spatial aggregation approach. The lodging area estimation models of dual-pol and quad-pol were established using a random forest (RF) algorithm. The Sobol approach evaluated the uncertainty and sensitivity at a regional scale. Finally, we analyzed the scattering mechanisms of the lodging field. Results indicate that the proposed method achieves the high performance of the lodging area estimates at the regional scale, in the testing set, with R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> and RMSE of the GF-3 model being 0.57 and 18.63%, and the Sentinel-1 model is 0.49 and 20.59%. Besides, the uncertainties of models are below 10%, and are insensitive to the variation of parameters inter-correlation. The depolarization effect and scattering randomness gradually weaken with the increase of lodging percentages. In contrast, the surface scattering quickly increases and finally dominates the total scattering after lodging percentages greater than 80%. This proposed approach would help develop a real-time crop lodging monitoring system using SAR data.

Self-supervised deep reconstruction of mixed strip-shredded text documents

The reconstruction of shredded documents consists of coherently arranging fragments of paper (shreds) to recover the original document(s). A great challenge in computational reconstruction is to properly evaluate the compatibility between the shreds. While traditional pixel-based approaches are not robust to real shredding, more sophisticated solutions compromise significantly time performance. The solution presented in this work extends our previous deep learning method for single-page reconstruction to a more realistic/complex scenario: the reconstruction of several mixed shredded documents at once. In our approach, the compatibility evaluation is modeled as a two-class (valid or invalid) pattern recognition problem. The model is trained in a self-supervised manner on samples extracted from simulated-shredded documents, which obviates manual annotation. Experimental results on three datasets – including a new collection of 100 strip-shredded documents produced for this work – have shown that the proposed method outperforms the competing ones on complex scenarios, achieving accuracy superior to 90%.

Can more carbon be captured by grasslands? A case study of Inner Mongolia, China

Grasslands cover a large part of the Earth's surface and play an important role in the global carbon cycle. Previous studies have indicated that nearly half of the grassland vegetation cover has experienced degradation on a global scale; if this degradation is reversed, grasslands can act as potential carbon sinks. However, the question of how much more carbon (carbon gap) could be sequestrated by grassland vegetation by regulating human activities remains unanswered. Here, we present an innovative approach to assess the achievable carbon gap through focal analysis of long-term Moderate Resolution Imaging Spectroradiometer (MODIS) Net Primary Production (NPP) dataset or observed NPP (ONPP). In focal analysis, region segmentation was done to produce spatially homogeneous patches of the same types of soil, topography, and vegetation, referred to as S-T-V units, to minimize the variation in environmental conditions and their impacts on the NPP. Then, the ONPP within each S-T-V unit was rectified by offsetting the variations in potential NPP determined by the climate-oriented Miami NPP model. Hence, spatial variations in the climate-rectified ONPP (ONPPCR) in an S-T-V unit were solely determined by different human activities across locations. In a case study of the Inner Mongolia grassland of China, three focal statistics, namely mean (Mean), 95% percentile threshold (95%PCT), and maximum (Max) within each S-T-V unit were computed for ONPPCR for each year from 2000 to 2014 to assess the annual carbon uptake that was achievable by updating grassland management practices. The carbon gaps were assessed to be 11.8, 58.9, and 74.6 gC/m2 per year based on Mean, 95%PCT, and Max, respectively, compared to 65.0 gC/m2 per year based on the traditional pixel-based approach. We conclude that the carbon gap patterns identified from focal analysis are practically achievable and are more valuable in formulating policy-related decisions for grassland management. Implementing sustainable management practices that are currently being practiced at locations with high ONPPCR in neighboring degraded areas is expected to increase the carbon sequestration by grassland vegetation by one-third.

A Parametric Level set Method for Imaging Multiphase Conductivity Using Electrical Impedance Tomography

Electrical impedance tomography (EIT) is an imaging modality that provides cross-sectional images of objects that carry contrasts in electrical conductivity. EIT is suitable for example for monitoring industrial processes involving multiple phases with different conductivities. This paper presents a parametric level set (PLS) based reconstruction scheme for an EIT-imaging of conductivity distributions within multiphase systems. The proposed scheme involves applying a multiphase level set model to solve the inverse problem of finding interfaces between the regions having different conductivity values and estimating the conductivities of phases. The unknown conductivity to be reconstructed is assumed to be piecewise constant while the interface between the regions are represented by the two PLS functions employing Gaussian radial basis functions (GRBF). The level set-based scheme handles topological merging and breaking naturally during the evolution process. It also offers several advantages compared to the traditional pixel-based approaches. For example, the representation of the PLS function by using GRBF provides flexibility in describing a large class of shapes with fewer unknowns. Numerical simulations and phantom experiments are performed to validate the proposed method.

Dimensionality Reduction and Classification of Hyperspectral Images Using Object-Based Image Analysis

Object-based image analysis (OBIA) has attained great importance for the delineation of landscape features, particularly with the accessibility to satellite images with high spatial resolution acquired by recent sensors. Statistical parametric classifiers have become ineffective mainly due to their assumption of normal distribution, vast increase in the dimensions of the data and availability of limited ground sample data. Despite pixel-based approaches, OBIA takes semantic information of extracted image objects into consideration, and thus provides more comprehensive image analysis. In this study, Indian Pines hyperspectral data set, which was recorded by the AVIRIS hyperspectral sensor, was used to analyse the effects of high dimensional data with limited ground reference data. To avoid the dimensionality curse, principal component analysis (PCA) and feature selection based on Jeffries–Matusita (JM) distance were utilized. First 19 principal components representing 98.5% of the image were selected using the PCA technique whilst 30 spectral bands of the image were determined using JM distance. Nearest neighbour (NN) and random forest (RF) classifiers were employed to test the performances of pixel- and object-based classification using conventional accuracy metrics. It was found that object-based approach outperformed the traditional pixel-based approach for all cases (up to 18% improvement). Also, the RF classifier produced significantly more accurate results (up to 10%) than the NN classifier.

Conservation-induced resettlement as a driver of land cover change in India: An object-based trend analysis

Located in the foothills of the Indian Himalaya, Rajaji National Park was established to protect and enhance the habitat of the Asian elephant (Elephas maximus) and tiger (Panthera tigris). In 2002 the Van Gujjars, indigenous forest pastoralists, were voluntarily resettled from the Chilla Range (an administrative unit of Rajaji National Park) to Gaindikhata, a nearby area where they were granted land for agriculture. In this study we used a variety of remote sensing approaches to identify changes in land cover associated with the resettlement. The methods comprise two main approaches. First, we used object-based image analysis (OBIA) to identify the pre-resettlement land cover classes of use areas (representing agricultural expansion and adjacent areas of grazing, and collection of fuelwood and fodder) and recovery areas (representing areas where settlements were removed, and the adjacent areas of resource use). Secondly, we used trend analysis to assess the gradual and abrupt changes in vegetation that took place in use and recovery areas. To conduct the trend analysis we used BFAST (Breaks For Additive Season and Trend), which separates seasonal variation from long-term trends, and identifies breaks that can be linked back to disturbances or land cover changes. We found that the OBIA classification yielded high average class accuracies, and we were able to make class distinctions that would have been difficult to make using a traditional pixel-based approach. Pre-resettlement, the recovery areas were classified as mixed forest and riparian vegetation. In contrast, the use areas were classified primarily as grass dominated, brush dominated, and plantation forest, and were located relatively far away from riparian areas. Following the resettlement, the trend analysis showed a sudden change in the seasonal variation of NDVI in areas converted to agriculture. Areas neighboring the new agricultural land experienced sudden decreases in NDVI, suggestive of disturbances, at a higher rate than the same land cover types elsewhere. At the same time, these neighboring areas experienced a gradual overall increase in NDVI which could be caused by an expansion of leafy invasive shrubs such as Lantana camara in areas heavily used for biomass collection. The recovery areas also experienced a gradual increase in NDVI as well as sudden breaks to this trend, but we lacked evidence to connect these changes to the resettlement. Our findings support the claim that the resettlement has shifted pressure from more ecologically valuable to less ecologically valuable land cover types, and suggest that to some degree resource use pressure has increased outside the park. The study employs a novel synthesis of OBIA and trend analysis that could be applied to land change studies more broadly.

Object-Oriented Approach of Landsat Imagery for Flood Mapping

study introduced an object-oriented approach to flood mapping and affected field estimation in central Cambodia. Traditional pixel-based image algorithms for flood mapping and land use and land cover classification endure from low accuracy, sub-pixel problems, and the cover noise effect in the resulting images On the other hand, the object-based image analysis (OBIA) approach has been thoroughly developed in the last two decades to overcome the limitations and disadvantages of the traditional pixel-based approaches by generating and analyzing meaningful image objects instead of individual pixels and reducing the speckle noise effect. The OBIA approach was applied for the image classification with a new improved estimation algorithm with multi scale parameter in the segmentation process to obtain more accurate results in the flood mapping. Flooding can be recognized using a variety of approaches such as statistics, ground-based measuring, prediction model, remote sensing techniques.

Multi-phase flow monitoring with electrical impedance tomography using level set based method

In this paper, a level set-based reconstruction scheme is applied to multi-phase flow monitoring using electrical impedance tomography (EIT). The proposed scheme involves applying a narrowband level set method to solve the inverse problem of finding the interface between the regions having different conductivity values. The multi-phase level set model for the conductivity distribution inside the domain is represented by two level set functions. The key principle of the level set-based method is to implicitly represent the shape of interface as the zero level set of higher dimensional function and then solve a set of partial differential equations. The level set-based scheme handles topological merging and breaking naturally during the evolution process. It also offers several advantages compared to traditional pixel-based approach. Level set-based method for multi-phase flow is tested with numerical and experimental data. It is found that level set-based method has better reconstruction performance when compared to pixel-based method.

Automatic Selection of Optimal Segmentation Scale of High-resolution Remote Sensing Images

With the increasing of spatial resolution of imaging sensors, object-oriented feature information extraction technology is developing rapidly. The advantages of object-based classification over the traditional pixel-based approach are well documented. Image segmentation is a key step to realize the object-oriented classification. The choice of scale parameter is very important and has a great influence on the segmentation effectiveness,but the choice of scale parameter is still decided by the repeated attempts and subjective judgments of operator,which are lacking in stability and reliability. Thus, an objective and unsupervised method is proposed for selecting optimal parameter for image segmentation to ensure best quality results. In this paper, WorldView 2 as data source, a new method based on principal component transform is introduced to choose an optimal parameter for image segmentation. We choose principal component images as the editor of image segmentation and eigenvalues as the weights of heterogeneity f and segmentation global score. Segmentation images, ranging from 20 to200 scale, step by 10, are created in Definiens Professional 8.7. Then, the global intra-segment and inter-segment heterogeneity indexes are taken into account to identify the optimal segmentation scale(i. e. the highest GS value) by using the cubic spline interpolation function method. After comparison with the results of image segmentation based on traditional three bands, image segmentation effect obtained by principal component transform has obvious advantages. As a result, the method in this paper can effectively avoid the subjectivity of the artificial segmentation, one-sidedness and inefficiency, improve the quality of high-resolution image segmentation.The method also makes a good preprocessing work for later image classification and information extraction.

Comparing ten classification methods for burned area mapping in a Mediterranean environment using Landsat TM satellite data

Various methods have been developed during the past three decades to improve the classification accuracy in burned area mapping using satellite data captured by different sensors. In this article, we compare ten such classification approaches using Landsat Thematic Mapper (TM) imagery on three Mediterranean test sites by evaluating the classification accuracy using (i) a traditional pixel-based approach, (ii) the concept of the Pareto boundary of efficient solution and (iii) linear regression analysis. Additionally, we make a discrimination of errors depending on their distribution and causal factor. The classification approaches compared resulted in not statistically significant differences in the accuracy of the burned area maps. Differences between the methods were also observed when considering the accuracy along the edges of the burned patches; however, again these were not statistically significant. The findings of our study in a Mediterranean environment clearly demonstrate that, for the selection of the most suitable classification approach, other factors could be given more weight, such as computational resources, imagery characteristics, availability of ancillary data, available software and the analyst's experience. Maybe the most important finding of our work is that the variance imposed by the methods is less than the variance imposed by factors differentiated locally in the three study sites since the between-group variance of the overall accuracy is higher than that of the within groups.

A multiscale geographic object-based image analysis to estimate lidar-measured forest canopy height using Quickbird imagery

Lidar (light detection and ranging) has demonstrated the ability to provide highly accurate information on forest vertical structure; however, lidar data collection and processing are still expensive. Very high spatial resolution optical remotely sensed data have also shown promising results to delineate various forest biophysical properties. In this study, our main objective is to examine the potential of Quickbird (QB) imagery to accurately estimate forest canopy heights measured from small-footprint lidar data. To achieve this, we have developed multiscale geographic object-based image analysis (GEOBIA) models from QB data for both deciduous and conifer stands. In addition to the spectral information, these models also included (1) image-texture [i.e., an internal-object variability measure and a new dynamic geographic object-based texture (GEOTEX) measure that quantifies forest variability within neighboring objects] and (2) a canopy shadow fraction measure that acts as a proxy of vertical forest structure. A novel object area-weighted error calculation approach was used to evaluate model performance by considering the importance of object size. To determine the best object scale [i.e., mean object size (MOS)] for defining the most accurate canopy height estimates, we introduce a new perspective, which considers height variability both between- and within-objects at all scales. To better evaluate the improvements resulting from our GEOBIA models, we compared their performance with a traditional pixel-based approach. Our results show that (1) the addition of image-texture and shadow fraction variables increases the model performance versus using spectral information only, especially for deciduous trees, where the average increase of R 2 is approximately 23% with a further 1.47 m decrease of Root Mean Squared Error (RMSE) at all scales using the GEOBIA approach; (2) the best object scale for our study site corresponds to an MOS of 4.00 ha; (3) at most scales, GEOBIA models achieve more accurate results than pixel-based models; however, we note that inappropriately selected object scales may result in poorer height accuracies than those derived from the applied pixel-based approach.

Assessing object-based classification: advantages and limitations

The advantages of object-based classification over the traditional pixel-based approach are well documented. However, the potential limitations of object-based classification remain less explored. In this letter, we assess the advantages and limitations of an object-based approach to remote sensing image classification relative to a pixel-based approach. We first quantified the negative impacts of under-segmentation errors on the potential accuracy of object-based classification by developing a new segmentation accuracy measure. Then we evaluated the advantages and limitations of object-based classification by quantifying their overall effects relative to pixel-based classification, with respect to their classification units and features at multiple segmentation scales. The results based on a QuickBird satellite image indicate that (1) segmentation accuracies decrease with increasing segmentation scales and the negative impacts of under-segmentation errors become significantly large at large scales and (2) there are both advantages and limitations in using object-based classification, and their trade-off determines the overall effect of object-based classification, which is dependent on the segmentation scales.

Mapping urban and peri-urban agriculture using high spatial resolution satellite data

In rapidly changing peri-urban environments where biophysical and socio- economic processes lead to spatial fragmentation of agricultural land, remote sensing offers an efficient tool to collect land cover/land use (LCLU) data for decision-making. Compared to traditional pixel-based approaches, remote sensing with object-based classification methods is reported to achieve improved classification results in complex heterogeneous landscapes. This study assessed the usefulness of object-oriented analysis of Quickbird high spatial resolution satellite data to classify urban and peri-urban agriculture in a limited peri-urban area of Hanoi, Vietnam. The results revealed that segmentation was essential in developing the object-oriented classification approach. Accurate segmentation of shape and size of an object enhanced classification with spectral, textural, morphological, and topological features. A qualitative, visual comparison of the classification results showed successful localisation and identification of most LCLU classes. Quantitative evaluation was conducted with a classification error matrix reaching an overall accuracy of 67% and a kappa coefficient of 0.61. In general, object-oriented classification of high spatial resolution satellite data proved the promising approach for LCLU analysis at village level. Capturing small-scale urban and peri-urban agricultural diversity offers a considerable potential for environmental monitoring. Challenges remain with the delineation of field boundaries and LCLU diversity on more spatially extensive datasets.

Microwave Imaging for Early Breast Cancer Detection Using a Shape-based Strategy

In this paper, we propose and analyze a novel shape reconstruction technique for the early detection of breast cancer from microwave data, which is based on a level-set technique. The shape-based approach offers several advantages compared to more traditional pixel-based approaches when targeting the reconstruction of key characteristics of a hidden tumor such as its correct size, shape, and static permittivity value. In addition to these key characteristics of hidden tumors, we aim at estimating the correct interfaces between fatty and fibroglandular tissue in the breast and their internal permittivity profiles. The level set strategy (which is an implicit representation of the shapes) frees us from topological restrictions when reconstructing an a priori arbitrary number of tumors and the often quite complicated interfaces between fatty and fibroglandular regions. The presented strategy is able to detect and, in most cases, characterize tumors whose sizes (diameters) are much smaller than the wavelengths of the electromagnetic waves that are used for illuminating the breast. We present numerical results for a 2-D model with two distinct tissue types (fatty and fibroglandular) in the interior of the breast (in addition to a possible tumor and the surrounding skin). Our results demonstrate the performance and potential of our scheme in various simulated but realistic situations.

QuickBird and Hyperion data analysis of an invasive plant species in the Galapagos Islands of Ecuador: Implications for control and land use management

In the Galapagos Islands of Ecuador, one of the greatest threats to the terrestrial ecosystem is the increasing number and areal extent of invasive species. Increased human presence on the islands has hastened the introduction of plant and animal species that threaten the native and endemic flora and fauna. Considerable research on invasive species in the Galapagos Islands has been conducted by the Charles Darwin Foundation. We complement that work through a spatially- and spectrally-explicit satellite assessment of an important invasive plant species (Psidium guajava — guava) on Isabela Island that integrates diverse remote sensing systems, data types, spatial and spectral resolutions, and analytical and image processing approaches. QuickBird and Hyperion satellite data are processed to characterize the areal extent and spatial structure of guava through the following approaches: (1) QuickBird data are classified through a traditional pixel-based approach (i.e., an unsupervised classification approach using the ISODATA algorithm), as well as an Object-Based Image Analysis (OBIA) approach; (2) multiple approaches for spectral “unmixing” of the Hyperion hyper-spectral data are assessed to construct spectral end-members from QuickBird data using linear and non-linear mixture modeling approaches; and (3) landscape pattern metrics are calculated and compared for the pixel-based, object-based, and spectral unmixing approaches. The spectral–spatial characteristics of guava are interpreted relative to management strategies for the control of guava and the restoration of natural ecosystems in the Galapagos National Park.

A level set evolution strategy in microwave imaging for early breast cancer detection

In this paper we propose and analyse a novel shape-reconstruction technique for the early detection of breast cancer from microwave data, which is based on a level-set technique. The shape-based approach offers several advantages compared to more traditional pixel-based approaches, as, for example, well-defined boundaries and the incorporation of an intrinsic regularization (in form of a-priori assumptions regarding the general anatomical structures present in the medium) that reduces the dimensionality of the inverse problem and thereby stabilizing the reconstruction. The level set strategy (which is an implicit representation of the shapes) frees us from topological restrictions during this reconstruction process. We present numerical results in 2D which demonstrate the performance of our scheme in various simulated realistic situations.

Object-based Detailed Vegetation Classification with Airborne High Spatial Resolution Remote Sensing Imagery

In this paper, we evaluate the capability of the high spatial resolution airborne Digital Airborne Imaging System (DAIS) imagery for detailed vegetation classification at the alliance level with the aid of ancillary topographic data. Image objects as minimum classification units were generated through the Fractal Net Evolution Approach (FNEA) segmentation using eCognition software. For each object, 52 features were calculated including spectral features, textures, topographic features, and geometric features. After statistically ranking the importance of these features with the classification and regression tree algorithm (CART), the most effective features for classification were used to classify the vegetation. Due to the uneven sample size for each class, we chose a non-parametric (nearest neighbor) classifier. We built a hierarchical classification scheme and selected features for each of the broadest categories to carry out the detailed classification, which significantly improved the accuracy. Pixel-based maximum likelihood classification (MLC) with comparable features was used as a benchmark in evaluating our approach. The objectbased classification approach overcame the problem of saltand-pepper effects found in classification results from traditional pixel-based approaches. The method takes advantage of the rich amount of local spatial information present in the irregularly shaped objects in an image. This classification approach was successfully tested at Point Reyes National Seashore in Northern California to create a comprehensive vegetation inventory. Computer-assisted classification of high spatial resolution remotely sensed imagery has good potential to substitute or augment the present ground-based inventory of National Park lands.

Object-oriented classification of Landsat ETM+ satellite image

This paper presents results of object-oriented classification of Landsat ETM+ satellite image conducted using eCognition software. The classified image was acquired on 7 May 2000 . In this particular study, an area of 423 km within the borders of Legionowo Community near Warsaw is considered. Prior to classification, segmentation of the Landsat ETM+ image is performed using panchromatic channel, fused multispectral and panchromatic data. The applied methods of classification enabled the identification of 18 land cover and land use classes. After the classification, generalization and raster to vector conversion, verification and accuracy assessment are performed by means of visual interpretation. Overall accuracy of the classification reached 94.6%. The verification and classification results are combined to form the final database. This is followed by comparing the object-oriented with traditional pixel-based classification. The latter is performed using the so-called hybrid classification based on both supervised and unsupervised classification approaches. The traditional pixel-based approach identified only 8 classes. Comparison of the pixel-based classification with the database obtained using the object-oriented approach revealed that the former reached 72% and 61% accuracy, according to the applied method.

Object-based retrieval of biophysical canopy variables using artificial neural nets and radiative transfer models

The retrieval of biophysical variables using canopy reflectance models is hampered by the fact that the inverse problem is ill-posed. This leads to unstable and often inaccurate inversion results. In order to regularize the model inversion, a novel approach has been developed and tested on synthetic Landsat TM reflectance data. The method takes into account the neighbouring radiometric information of the pixel of interest, named object signature. The neighbourhood data can either be extracted from gliding windows, already segmented images, or using digitized field boundaries. The extracted radiometric data of the neighbourhood pixels are used to calculate 42 descriptive statistical properties that comprehensively characterize the spectral (co)variance of the image object (e.g. mean and standard deviation of the distributions, intercorrelations between spectral bands, etc.). Together with the habitual spectral signature of the pixel being inverted (6 variables), this object signature (42 variables) is used as input in an artificial neural net to estimate simultaneously three important biophysical variables (i.e. leaf area index, leaf chlorophyll, and leaf water content). The use of neural nets for the model inversion avoids time-consuming iterative optimizations and provides a computational effective way to consider simultaneously pixel and object signatures. In order to “learn” the relation between spectral signatures and biophysical variables, the neural nets were previously trained on large synthetic data sets. The data sets consist of pixel signatures and the corresponding signatures of image objects representing various agricultural fields. The signatures were simulated with the SAILH+PROSPECT canopy reflectance model, assuming largely varying intra- and interfield distributions of the model input parameters. To demonstrate the benefits of the object-based inversion, neural nets were also trained on the pixel signatures alone. For this purpose, the object signatures were simply replaced by randomly generated white noise—all other conditions being the same. The intercomparison based on 30,000 independent validation patterns showed that the proposed method significantly enhances the estimation accuracies: for example, the leaf area index (LAI) is estimated with a percental root mean square error (PRMSE) of 18.3% (object-based) compared to 25.1% (pixel based); the corresponding numbers for the leaf chlorophyll content are 12.6% compared to 15.9%; for the equivalent leaf water thickness, 10.6% and 13.9%, respectively. The benefit of the object signature was strongest for the LAI. Concerning this important biophysical variable, the novel concept accounted for almost one-half of the remaining unexplained variance of the traditional pixel-based approach. Increased accuracies were attributed to the fact that intrafield variations of biophysical canopy variables lead to object signatures that are modulated by the actual average leaf angle (ALA) of the canopy. Since ALA can be considered constant insight given an agricultural field, the concurrent use of pixel and object signatures significantly reduces confounding effects between LAI and ALA typical for traditional inversion approaches. Compared to competing approaches, the algorithm can also be applied to monotemporal imagery and does not require a priori information or the identification of crop type.