The authors present a qualitative and quantitative comparison of various similarity measures that form the kernel of common area-based stereo-matching systems. The authors compare classical difference and correlation measures as well as nonparametric measures based on the rank and census transforms for a number of outdoor images. For robotic applications, important considerations include robustness to image defects such as intensity variation and noise, the number of false matches, and computational complexity. In the absence of ground truth data, the authors compare the matching techniques based on the percentage of matches that pass the left-right consistency test. The authors also evaluate the discriminatory power of several match validity measures that are reported in the literature for eliminating false matches and for estimating match confidence. For guidance applications, it is essential to have and estimate of confidence in the three-dimensional points generated by stereo vision. Finally, a new validity measure, the rank constraint, is introduced that is capable of resolving ambiguous matches for rank transform-based matching.

Quantitative evaluation of matching methods and validity measures for stereo vision

Deep learning is transforming most areas of science and technology, including electron microscopy. This review paper offers a practical perspective aimed at developers with limited familiarity. For context, we review popular applications of deep learning in electron microscopy. Following, we discuss hardware and software needed to get started with deep learning and interface with electron microscopes. We then review neural network components, popular architectures, and their optimization. Finally, we discuss future directions of deep learning in electron microscopy.

Review: Deep Learning in Electron Microscopy

Optical remotely sensed data are typically discontinuous, with missing values due to cloud cover. Consequently, gap-filling solutions are needed for accurate crop phenology characterization. The here presented Decomposition and Analysis of Time Series software (DATimeS) expands established time series interpolation methods with a diversity of advanced machine learning fitting algorithms (e.g., Gaussian Process Regression: GPR) particularly effective for the reconstruction of multiple-seasons vegetation temporal patterns. DATimeS is freely available as a powerful image time series software that generates cloud-free composite maps and captures seasonal vegetation dynamics from regular or irregular satellite time series. This work describes the main features of DATimeS, and provides a demonstration case using Sentinel-2 Leaf Area Index time series data over a Spanish site. GPR resulted as an optimum fitting algorithm with most accurate gap-filling performance and associated uncertainties. DATimeS further quantified LAI fluctuations among multiple crop seasons and provided phenological indicators for specific crop types.

DATimeS: A machine learning time series GUI toolbox for gap-filling and vegetation phenology trends detection

Understanding the impacts of drought and climate change on vegetation dynamics is of great significance in terms of formulating vegetation management strategies and predicting future vegetation growth. In this study, Pearson correlation analysis was used to investigate the correlations between drought, climatic factors and vegetation conditions, and linear regression analysis was adopted to investigate the time-lag and time-accumulation effects of climatic factors on vegetation coverage based on the standardized evapotranspiration deficit index (SEDI), normalized difference vegetation index (NDVI), and gridded meteorological dataset in the Yellow River Basin (YLRB) and Yangtze River Basin (YTRB), China. The results showed that (1) the SEDI in the YLRB showed no significant change over time and space during the growing season from 1982 to 2015, whereas it increased significantly in the YTRB (slope = 0.013/year, p < 0.01), and more than 40% of the area showed a significant trend of wetness. The NDVI of the two basins, YLRB and YTRB, increased significantly at rate of 0.011/decade and 0.016/decade, respectively (p < 0.01). (2) Drought had a significant impact on vegetation in 49% of the YLRB area, which was mainly located in the northern region. In the YTRB, the area significantly affected by drought accounted for 21% of the total area, which was mainly distributed in the Sichuan Basin. (3) In the YLRB, both temperature and precipitation generally had a one-month accumulated effect on vegetation conditions, while in the YTRB, temperature was the major factor leading to changes in vegetation. In most of the area of the YTRB, the effect of temperature on vegetation was also a one-month accumulated effect, but there was no time effect in the Sichuan Basin. Considering the time effects, the contribution of climatic factors to vegetation change in the YLRB and YTRB was 76.7% and 63.2%, respectively. The explanatory power of different vegetation types in the two basins both increased by 2% to 6%. The time-accumulation effect of climatic factors had a stronger explanatory power for vegetation growth than the time-lag effect.

/pdf/impacts-of-drought-and-climatic-factors-on-vegetation-veltrrlf.pdf

Impacts of Drought and Climatic Factors on Vegetation Dynamics in the Yellow River Basin and Yangtze River Basin, China

The Near Surface Concentrations (NSC) of O3, CO, and NO2 are crucial worldwide indicators of air quality However, current frameworks devised for the estimation of the NSC of O3, CO, and NO2 have defects, such as coarse spatial resolution and large missing coverage To address this issue, this study aims to estimate the daily (~13:30 local time) full-coverage NSC of O3, CO, and NO2 at a high spatial resolution (005° for O3 and NO2; 007° for CO) over China by using datasets from S5P-TROPOMI and GEOS-FP In specific, the light gradient boosting machine is employed to train the estimation models Validation results show that the NSC of O3, CO, and NO2 are well estimated, with the R2s of 091, 071, and 083 for the sample-based cross validation, respectively Meanwhile, the proposed framework achieves a satisfactory performance in comparison to the latest related works, as reflected by the estimation accuracy and spatial resolution As for the mapping, the estimated results show coherent spatial distribution and can accurately grasp the seasonal characteristics of each air pollutant Finally, the estimated results are utilized to analyze the temporal variations of O3, CO, and NO2 during the COrona VIrus Disease 2019 (COVID-19) lockdown in China, which is an extend application for adopting the proposed framework in air quality monitoring Results show that the estimated NSC of O3, CO, and NO2 in 2020 present significant variations during different periods of the COVID-19 lockdown in China compared to last year In addition, the variations in the NSC of O3, CO, and NO2 during the COVID-19 lockdown in China possibly result from restrictions in the anthropogenic activities

Estimating daily full-coverage near surface O3, CO, and NO2 concentrations at a high spatial resolution over China based on S5P-TROPOMI and GEOS-FP

This paper presents the nearest neighbor value (NNV) algorithm for high resolution (H.R.) image interpolation. The difference between the proposed algorithm and conventional nearest neighbor algorithm is that the concept applied, to estimate the missing pixel value, is guided by the nearest value rather than the distance. In other words, the proposed concept selects one pixel, among four directly surrounding the empty location, whose value is almost equal to the value generated by the conventional bilinear interpolation algorithm. The proposed method demonstrated higher performances in terms of H.R. when compared to the conventional interpolation algorithms mentioned.

/pdf/nearest-neighbor-value-interpolation-2kaoj8a414.pdf

Nearest neighbor value interpolation

This paper proposes an optimization scheme for the image interpolation algorithms, in particular the bilinear algorithm. The only original point is a decision step in which it is decided whether the four neighbouring pixels have the same value and if so the conventional bilinear interpolation is replaced by a nearest neighbour interpolation. The experimental results corroborated the efficiency of the proposed scheme over conventional bilinear and showed improvements in terms of speed and quality, especially in case where images with less grain textures have been interpolated.

Optimization of image interpolation based on nearest neighbour algorithm

Rescaling bilinear (RB) interpolant's pixels is a novel image interpolation scheme. In the current study, we investigate the effects on the quality of interpolated images. RB determines the lower and upper bounds using the standard deviation of the four nearest pixels to find the new interval or range that will be used to rescale the bilinear interpolant's pixels. The products of the rescaled-pixels and corresponding distance-based-weights are added to estimate the new pixel value, to be assigned at the empty locations of the destination image. Effects of RB on image interpolation quality were investigated using standard full-reference and non-reference objective image quality metrics, particularly those focusing on interpolated images features and distortion similarities. Furthermore, variance and mean based metrics were also employed to further investigate the effects in terms of contrast and intensity increment or decrement. The Matlab based simulations demonstrated generally superior performances of RB compared to the traditional bilinear (TB) interpolation algorithm. The studied scheme's major drawback was a higher processing time and tendency to rely on the image type and/or specific interpolation scaling ratio to achieve superior performances. Potential applications of rescaling based bilinear interpolation may also include ultrasound scan conversion in cardiac ultrasound, endoscopic ultrasound, etc. (Less)

/pdf/effects-of-rescaling-bilinear-interpolant-on-image-pzhe44r9cz.pdf

Effects of rescaling bilinear interpolant on image interpolation quality

This paper proposes a novel method optimizing the bilinear interpolation using Ant Colony Algorithm (ACA). The isotropous-assignment of the old-pixels’ average-value to an unknown point (or pixel), in bilinear interpolation, is responsible for many artifacts which corrupt the resulting image features, edges in particular. To correct, we apply the ant colony algorithm to find a reasonable value from all the directions in the neighborhoods of the unknown point. The experimental results show better immanent image features than bilinear interpolation.

Optimization of Bilinear Interpolation Based on Ant Colony Algorithm

This paper proposes a novel algorithm for image interpolation. The motivation is based on relentless image interpolation artefacts and computational complexity. The solution proposed is founded on reprocessing one of the four pixels surrounding the unknown location and calculating the mean between that pixel and the value introduced by the bilinear interpolation. Subsequently, the mean is multiplied by the control factor k whose value is selected according to experimental analysis. Experimental results are further presented to show the effectiveness and performance of the proposed algorithm.

Olivier Rukundo

Papers

Nearest neighbor value interpolation

Optimization of image interpolation based on nearest neighbour algorithm

Effects of rescaling bilinear interpolant on image interpolation quality

Optimization of Bilinear Interpolation Based on Ant Colony Algorithm

Image interpolation based on the pixel value corresponding to the smallest absolute difference