Search

Plant ecology and biodiversity research have increasingly incorporated trait-based approaches and remote sensing. Compared with traditional field survey (which typically samples individual trees), remote sensing enables quantifying functional traits over large contiguous areas, but assigning trait values to biological units such as species and individuals is difficult with pixel-based approaches. We used a subtropical forest landscape in China to compare an approach based on airborne LiDAR-delineated individual tree crowns (ITCs) with a pixel-based approach for assessing functional traits from remote sensing data. We compared trait distributions, trait–trait relationships and functional diversity metrics obtained by the ITC- and pixel-based approaches at changing pixel size and extent. We found that morphological traits derived from airborne laser scanning showed more differences between ITC- and pixel-based approaches than physiological traits estimated by airborne Pushbroom Hyperspectral Imager-3 (PHI-3) hyperspectral data. Pixel sizes approximating average tree crowns yielded similar results as ITCs, but 95th quantile height and foliage height diversity tended to be overestimated and leaf area index underestimated relative to ITC-based values. With increasing pixel size, the differences to ITC-based trait values became larger and less trait variance was captured, indicating information loss. The consistency of ITC- and pixel-based functional richness also decreased with increasing pixel size, and changed with the observed extent for functional diversity monitoring. We conclude that whereas ITC-based approaches in principle allow partitioning of variation between individuals, genotypes and species, high-resolution pixel-based approaches come close to this and can be suitable for assessing ecosystem-scale trait variation by weighting individuals and species according to coverage.

Pixel-based (PB) and geographic-object-based (GEOBIA) classification approaches allow the extraction of different objects from multispectral images (MS). The primary goal of this research was the analysis of UAV imagery applicability and accuracy assessment of MLC and SVM classification algorithms within PB and GEOBIA classification approaches. The secondary goal was to use different accuracy assessment metrics to determine which of the two tested classification algorithms (SVM and MLC) most reliably distinguishes olive tree crowns and which approach is more accurate (PB or GEOBIA). The third goal was to add false polygon samples for Correctness (COR), Completeness (COM) and Overall Quality (OQ) metrics and use them to calculate the Total Accuracy (TA). The methodology can be divided into six steps, from data acquisition to selection of the best classification algorithm after accuracy assessment. High-quality DOP (digital orthophoto) and UAVMS were generated. A new accuracy metric, called Total Accuracy (TA), combined both false and true positive polygon samples, thus providing a more comprehensive insight into the assessed classification accuracy. The SVM (GEOBIA) was the most reliable classification algorithm for extracting olive tree crowns from UAVMS imagery. The assessment carried out indicated that application of GEOBIA-SVM achieved a TACOR of 0.527, TACOM of 0.811, TAOQ of 0.745, Overall Accuracy (OA) of 0.926 or 0.980 and Area Under Curve (AUC) value of 0.904 or 0.929. The calculated accuracy metrics confirmed that the GEOBIA approach (SVM and MLC) achieved more accurate olive tree crown extraction than the PB approach (SVM and MLC) if applied to classifying VHR UAVMS imagery. The SVM classification algorithm extracted olive tree crowns more accurately than MLC in both approaches. However, the accuracy assessment has proven that PB classification algorithms can also achieve satisfactory accuracy.

Forests play a vital role in maintaining the global carbon balance. However, globally, forest ecosystems are increasingly threatened by climate change and deforestation in recent years. Monitoring forests, specifically forest biomass is essential for tracking changes in carbon stocks and the global carbon cycle. However, developing countries lack the capacity to actively monitor forest carbon stocks, which ultimately adds uncertainties in estimating country specific contribution to the global carbon emissions. In India, authorities use field-based measurements to estimate biomass, which becomes unfeasible to implement at finer scales due to higher costs. To address this, the present study proposed a framework to monitor above-ground biomass (AGB) at finer scales using open-source satellite data. The framework integrated four machine learning (ML) techniques with field surveys and satellite data to provide continuous spatial estimates of AGB at finer resolution. The application of this framework is exemplified as a case study for a dry deciduous tropical forest in India. The results revealed that for wet season Sentinel-2 satellite data, the Random Forest (adjusted R 2 = 0.91) and Artificial Neural Network (adjusted R 2 = 0.77) ML models were better-suited for estimating AGB in the study area. For dry season satellite data, all the ML models failed to estimate AGB adequately (adjusted R 2 between −0.05 – 0.43). Ensemble analysis of ML predictions not only made the results more reliable, but also quantified spatial uncertainty in the predictions as a metric to identify its robustness. [Display omitted] • Machine Learning techniques were used to estimate Above-Ground Biomass (AGB). • Wet season Sentinel-2 imagery was found sensitive to changes in AGB. • Multiple band combinations act as better predictor variables than single band. • Uncertainty map helps to identify regions with robust biomass prediction. • Easily replicable framework, applicable to other dry deciduous tropical forests. [ABSTRACT FROM AUTHOR]

In the current difficult time of COVID-19 pandemic, social distancing is a common practice adopted by healthy and infected people all over the world. Because of this, electronic healthcare and Telemedicine are coming up a big way. One of the most vital challenges in E-healthcare or telemedicine is the authentication and secure transmission of medical images received by an expert(doctor) at a far-off location from the sender(patient). To address the critical authentication issue, this paper proposes a blind pixel-based self-embedding fragile watermarking approach that is effective enough to localize the tampered region with more than 90% accuracy. The self-embedding approach plays a significant role in fragile watermarking for detecting the tampered region at the receiver side. Most of the self-embedding-based fragile watermarking approaches available in the literature are block-based which has a high False Positive Rate(FPR). In the proposed approach, every pixel of the image dynamically generates eight self-mutating offsprings and gets associated with them tightly at bit level. After a systematic set of procedures on that offsprings with a triple layer of security, a single authentication bit is selected using 16 × 1 multiplexer for each pixel intensity of the image. That is why the proposed approach has a high imperceptibility level (more than 60 dB PSNR). Experimental results confirm the efficacy of the proposed approach in terms of imperceptibility and tamper detection rate. [ABSTRACT FROM AUTHOR]

Pixel-based (PB) and geographic-object-based (GEOBIA) classification approaches allow the extraction of different objects from multispectral images (MS). The primary goal of this research was the analysis of UAV imagery applicability and accuracy assessment of MLC and SVM classification algorithms within PB and GEOBIA classification approaches. The secondary goal was to use different accuracy assessment metrics to determine which of the two tested classification algorithms (SVM and MLC) most reliably distinguishes olive tree crowns and which approach is more accurate (PB or GEOBIA). The third goal was to add false polygon samples for Correctness (COR), Completeness (COM) and Overall Quality (OQ) metrics and use them to calculate the Total Accuracy (TA). The methodology can be divided into six steps, from data acquisition to selection of the best classification algorithm after accuracy assessment. High-quality DOP (digital orthophoto) and UAVMS were generated. A new accuracy metric, called Total Accuracy (TA), combined both false and true positive polygon samples, thus providing a more comprehensive insight into the assessed classification accuracy. The SVM (GEOBIA) was the most reliable classification algorithm for extracting olive tree crowns from UAVMS imagery. The assessment carried out indicated that application of GEOBIA-SVM achieved a TACOR of 0.527, TACOM of 0.811, TAOQ of 0.745, Overall Accuracy (OA) of 0.926 or 0.980 and Area Under Curve (AUC) value of 0.904 or 0.929. The calculated accuracy metrics confirmed that the GEOBIA approach (SVM and MLC) achieved more accurate olive tree crown extraction than the PB approach (SVM and MLC) if applied to classifying VHR UAVMS imagery. The SVM classification algorithm extracted olive tree crowns more accurately than MLC in both approaches. However, the accuracy assessment has proven that PB classification algorithms can also achieve satisfactory accuracy. [ABSTRACT FROM AUTHOR]